{"cells": [{"cell_type": "markdown", "id": "22b293d7", "metadata": {"papermill": {"duration": 0.012549, "end_time": "2022-05-12T12:18:58.534467", "exception": false, "start_time": "2022-05-12T12:18:58.521918", "status": "completed"}, "tags": []}, "source": ["\n", "# Tutorial 9: Normalizing Flows for Image Modeling\n", "\n", "* **Author:** Phillip Lippe\n", "* **License:** CC BY-SA\n", "* **Generated:** 2022-05-12T13:44:24.674574\n", "\n", "In this tutorial, we will take a closer look at complex, deep normalizing flows.\n", "The most popular, current application of deep normalizing flows is to model datasets of images.\n", "As for other generative models, images are a good domain to start working on because\n", "(1) CNNs are widely studied and strong models exist,\n", "(2) images are high-dimensional and complex,\n", "and (3) images are discrete integers.\n", "In this tutorial, we will review current advances in normalizing flows for image modeling,\n", "and get hands-on experience on coding normalizing flows.\n", "Note that normalizing flows are commonly parameter heavy and therefore computationally expensive.\n", "We will use relatively simple and shallow flows to save computational cost and allow you to run the notebook on CPU,\n", "but keep in mind that a simple way to improve the scores of the flows we study here is to make them deeper.\n", "This notebook is part of a lecture series on Deep Learning at the University of Amsterdam.\n", "The full list of tutorials can be found at https://uvadlc-notebooks.rtfd.io.\n", "\n", "\n", "---\n", "Open in [![Open In Colab](data:image/png;base64,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){height=\"20px\" width=\"117px\"}](https://colab.research.google.com/github/PytorchLightning/lightning-tutorials/blob/publication/.notebooks/course_UvA-DL/09-normalizing-flows.ipynb)\n", "\n", "Give us a \u2b50 [on Github](https://www.github.com/PytorchLightning/pytorch-lightning/)\n", "| Check out [the documentation](https://pytorch-lightning.readthedocs.io/en/stable/)\n", "| Join us [on Slack](https://www.pytorchlightning.ai/community)"]}, {"cell_type": "markdown", "id": "79f5e211", "metadata": {"papermill": {"duration": 0.00964, "end_time": "2022-05-12T12:18:58.554488", "exception": false, "start_time": "2022-05-12T12:18:58.544848", "status": "completed"}, "tags": []}, "source": ["## Setup\n", "This notebook requires some packages besides pytorch-lightning."]}, {"cell_type": "code", "execution_count": 1, "id": "66ca3c42", "metadata": {"colab": {}, "colab_type": "code", "execution": {"iopub.execute_input": "2022-05-12T12:18:58.575803Z", "iopub.status.busy": "2022-05-12T12:18:58.575120Z", "iopub.status.idle": "2022-05-12T12:19:02.710809Z", "shell.execute_reply": "2022-05-12T12:19:02.709792Z"}, "id": "LfrJLKPFyhsK", "lines_to_next_cell": 0, "papermill": {"duration": 4.149137, "end_time": "2022-05-12T12:19:02.713186", "exception": false, "start_time": "2022-05-12T12:18:58.564049", "status": "completed"}, "tags": []}, "outputs": [], "source": ["! pip install --quiet \"ipython[notebook]\" \"seaborn\" \"tabulate\" \"torchvision\" \"setuptools==59.5.0\" \"pytorch-lightning>=1.4\" \"matplotlib\" \"torchmetrics>=0.7\" \"torch>=1.8\""]}, {"cell_type": "markdown", "id": "ba2a1738", "metadata": {"papermill": {"duration": 0.010226, "end_time": "2022-05-12T12:19:02.734320", "exception": false, "start_time": "2022-05-12T12:19:02.724094", "status": "completed"}, "tags": []}, "source": ["<div class=\"center-wrapper\"><div class=\"video-wrapper\"><iframe src=\"https://www.youtube.com/embed/U1fwesIusbg\" title=\"YouTube video player\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture\" allowfullscreen></iframe></div></div>\n", "Throughout this notebook, we make use of [PyTorch Lightning](https://pytorch-lightning.readthedocs.io/en/stable/).\n", "The first cell imports our usual libraries."]}, {"cell_type": "code", "execution_count": 2, "id": "6fb4c201", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:02.755979Z", "iopub.status.busy": "2022-05-12T12:19:02.755562Z", "iopub.status.idle": "2022-05-12T12:19:07.128426Z", "shell.execute_reply": "2022-05-12T12:19:07.127622Z"}, "papermill": {"duration": 4.386077, "end_time": "2022-05-12T12:19:07.130032", "exception": false, "start_time": "2022-05-12T12:19:02.743955", "status": "completed"}, "tags": []}, "outputs": [{"name": "stderr", "output_type": "stream", "text": ["/usr/lib/python3.8/site-packages/apex/pyprof/__init__.py:5: FutureWarning: pyprof will be removed by the end of June, 2022\n", "  warnings.warn(\"pyprof will be removed by the end of June, 2022\", FutureWarning)\n"]}, {"name": "stderr", "output_type": "stream", "text": ["/usr/local/lib/python3.8/dist-packages/numpy/core/getlimits.py:499: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.\n", "  setattr(self, word, getattr(machar, word).flat[0])\n", "/usr/local/lib/python3.8/dist-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float32'> type is zero.\n", "  return self._float_to_str(self.smallest_subnormal)\n", "/usr/local/lib/python3.8/dist-packages/numpy/core/getlimits.py:499: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.\n", "  setattr(self, word, getattr(machar, word).flat[0])\n", "/usr/local/lib/python3.8/dist-packages/numpy/core/getlimits.py:89: UserWarning: The value of the smallest subnormal for <class 'numpy.float64'> type is zero.\n", "  return self._float_to_str(self.smallest_subnormal)\n"]}, {"name": "stderr", "output_type": "stream", "text": ["WARNING:root:Bagua cannot detect bundled NCCL library, Bagua will try to use system NCCL instead. If you encounter any error, please run `import bagua_core; bagua_core.install_deps()` or the `bagua_install_deps.py` script to install bundled libraries.\n"]}, {"name": "stderr", "output_type": "stream", "text": ["/tmp/ipykernel_5302/3974796986.py:28: DeprecationWarning: `set_matplotlib_formats` is deprecated since IPython 7.23, directly use `matplotlib_inline.backend_inline.set_matplotlib_formats()`\n", "  set_matplotlib_formats(\"svg\", \"pdf\")  # For export\n", "Global seed set to 42\n"]}, {"name": "stdout", "output_type": "stream", "text": ["Using device cuda:0\n"]}], "source": ["import math\n", "import os\n", "import time\n", "import urllib.request\n", "from urllib.error import HTTPError\n", "\n", "import matplotlib\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "import pytorch_lightning as pl\n", "import seaborn as sns\n", "import tabulate\n", "import torch\n", "import torch.nn as nn\n", "import torch.nn.functional as F\n", "import torch.optim as optim\n", "import torch.utils.data as data\n", "import torchvision\n", "from IPython.display import HTML, display, set_matplotlib_formats\n", "from matplotlib.colors import to_rgb\n", "from pytorch_lightning.callbacks import LearningRateMonitor, ModelCheckpoint\n", "from torch import Tensor\n", "from torchvision import transforms\n", "from torchvision.datasets import MNIST\n", "from tqdm.notebook import tqdm\n", "\n", "%matplotlib inline\n", "set_matplotlib_formats(\"svg\", \"pdf\")  # For export\n", "matplotlib.rcParams[\"lines.linewidth\"] = 2.0\n", "sns.reset_orig()\n", "\n", "# Path to the folder where the datasets are/should be downloaded (e.g. MNIST)\n", "DATASET_PATH = os.environ.get(\"PATH_DATASETS\", \"data\")\n", "# Path to the folder where the pretrained models are saved\n", "CHECKPOINT_PATH = os.environ.get(\"PATH_CHECKPOINT\", \"saved_models/tutorial11\")\n", "\n", "# Setting the seed\n", "pl.seed_everything(42)\n", "\n", "# Ensure that all operations are deterministic on GPU (if used) for reproducibility\n", "torch.backends.cudnn.determinstic = True\n", "torch.backends.cudnn.benchmark = False\n", "\n", "# Fetching the device that will be used throughout this notebook\n", "device = torch.device(\"cpu\") if not torch.cuda.is_available() else torch.device(\"cuda:0\")\n", "print(\"Using device\", device)"]}, {"cell_type": "markdown", "id": "da0e0e19", "metadata": {"papermill": {"duration": 0.010289, "end_time": "2022-05-12T12:19:07.151414", "exception": false, "start_time": "2022-05-12T12:19:07.141125", "status": "completed"}, "tags": []}, "source": ["Again, we have a few pretrained models. We download them below to the specified path above."]}, {"cell_type": "code", "execution_count": 3, "id": "58b482d6", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:07.173932Z", "iopub.status.busy": "2022-05-12T12:19:07.173273Z", "iopub.status.idle": "2022-05-12T12:19:07.720437Z", "shell.execute_reply": "2022-05-12T12:19:07.719635Z"}, "papermill": {"duration": 0.561044, "end_time": "2022-05-12T12:19:07.722543", "exception": false, "start_time": "2022-05-12T12:19:07.161499", "status": "completed"}, "tags": []}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["Downloading https://raw.githubusercontent.com/phlippe/saved_models/main/tutorial11/MNISTFlow_simple.ckpt...\n", "Downloading https://raw.githubusercontent.com/phlippe/saved_models/main/tutorial11/MNISTFlow_vardeq.ckpt...\n"]}, {"name": "stdout", "output_type": "stream", "text": ["Downloading https://raw.githubusercontent.com/phlippe/saved_models/main/tutorial11/MNISTFlow_multiscale.ckpt...\n"]}], "source": ["# Github URL where saved models are stored for this tutorial\n", "base_url = \"https://raw.githubusercontent.com/phlippe/saved_models/main/tutorial11/\"\n", "# Files to download\n", "pretrained_files = [\"MNISTFlow_simple.ckpt\", \"MNISTFlow_vardeq.ckpt\", \"MNISTFlow_multiscale.ckpt\"]\n", "# Create checkpoint path if it doesn't exist yet\n", "os.makedirs(CHECKPOINT_PATH, exist_ok=True)\n", "\n", "# For each file, check whether it already exists. If not, try downloading it.\n", "for file_name in pretrained_files:\n", "    file_path = os.path.join(CHECKPOINT_PATH, file_name)\n", "    if not os.path.isfile(file_path):\n", "        file_url = base_url + file_name\n", "        print(\"Downloading %s...\" % file_url)\n", "        try:\n", "            urllib.request.urlretrieve(file_url, file_path)\n", "        except HTTPError as e:\n", "            print(\n", "                \"Something went wrong. Please try to download the file from the GDrive folder, or contact the author with the full output including the following error:\\n\",\n", "                e,\n", "            )"]}, {"cell_type": "markdown", "id": "153c5e1a", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.010051, "end_time": "2022-05-12T12:19:07.743999", "exception": false, "start_time": "2022-05-12T12:19:07.733948", "status": "completed"}, "tags": []}, "source": ["We will use the MNIST dataset in this notebook.\n", "MNIST constitutes, despite its simplicity, a challenge for small generative models as it requires the global understanding of an image.\n", "At the same time, we can easily judge whether generated images come from the same distribution as the dataset\n", "(i.e. represent real digits), or not.\n", "\n", "To deal better with the discrete nature of the images, we transform them\n", "from a range of 0-1 to a range of 0-255 as integers."]}, {"cell_type": "code", "execution_count": 4, "id": "b7465ad2", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:07.766358Z", "iopub.status.busy": "2022-05-12T12:19:07.765872Z", "iopub.status.idle": "2022-05-12T12:19:08.608429Z", "shell.execute_reply": "2022-05-12T12:19:08.607691Z"}, "papermill": {"duration": 0.855986, "end_time": "2022-05-12T12:19:08.610270", "exception": false, "start_time": "2022-05-12T12:19:07.754284", "status": "completed"}, "tags": []}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz\n", "Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz to /__w/1/s/.datasets/MNIST/raw/train-images-idx3-ubyte.gz\n"]}, {"data": {"application/vnd.jupyter.widget-view+json": {"model_id": "6512366631a949678ad0cfd766ffba0a", "version_major": 2, "version_minor": 0}, "text/plain": ["  0%|          | 0/9912422 [00:00<?, ?it/s]"]}, "metadata": {}, "output_type": "display_data"}, {"name": "stdout", "output_type": "stream", "text": ["Extracting /__w/1/s/.datasets/MNIST/raw/train-images-idx3-ubyte.gz to /__w/1/s/.datasets/MNIST/raw\n"]}, {"name": "stdout", "output_type": "stream", "text": ["\n", "Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz\n", "Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz to /__w/1/s/.datasets/MNIST/raw/train-labels-idx1-ubyte.gz\n"]}, {"data": {"application/vnd.jupyter.widget-view+json": {"model_id": "ce6bcc8f63c346b4a80e0c245eb89704", "version_major": 2, "version_minor": 0}, "text/plain": ["  0%|          | 0/28881 [00:00<?, ?it/s]"]}, "metadata": {}, "output_type": "display_data"}, {"name": "stdout", "output_type": "stream", "text": ["Extracting /__w/1/s/.datasets/MNIST/raw/train-labels-idx1-ubyte.gz to /__w/1/s/.datasets/MNIST/raw\n", "\n", "Downloading http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz\n", "Downloading http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz to /__w/1/s/.datasets/MNIST/raw/t10k-images-idx3-ubyte.gz\n"]}, {"data": {"application/vnd.jupyter.widget-view+json": {"model_id": "80a80e415bb24e459eef8028675f66ad", "version_major": 2, "version_minor": 0}, "text/plain": ["  0%|          | 0/1648877 [00:00<?, ?it/s]"]}, "metadata": {}, "output_type": "display_data"}, {"name": "stdout", "output_type": "stream", "text": ["Extracting /__w/1/s/.datasets/MNIST/raw/t10k-images-idx3-ubyte.gz to /__w/1/s/.datasets/MNIST/raw\n"]}, {"name": "stdout", "output_type": "stream", "text": ["\n", "Downloading http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz\n", "Downloading http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz to /__w/1/s/.datasets/MNIST/raw/t10k-labels-idx1-ubyte.gz\n"]}, {"data": {"application/vnd.jupyter.widget-view+json": {"model_id": "75f3db6cb8e14d1296d489ca26b01f24", "version_major": 2, "version_minor": 0}, "text/plain": ["  0%|          | 0/4542 [00:00<?, ?it/s]"]}, "metadata": {}, "output_type": "display_data"}, {"name": "stdout", "output_type": "stream", "text": ["Extracting /__w/1/s/.datasets/MNIST/raw/t10k-labels-idx1-ubyte.gz to /__w/1/s/.datasets/MNIST/raw\n", "\n"]}, {"name": "stderr", "output_type": "stream", "text": ["/usr/local/lib/python3.8/dist-packages/torchvision/datasets/mnist.py:498: UserWarning: The given NumPy array is not writeable, and PyTorch does not support non-writeable tensors. This means you can write to the underlying (supposedly non-writeable) NumPy array using the tensor. You may want to copy the array to protect its data or make it writeable before converting it to a tensor. This type of warning will be suppressed for the rest of this program. (Triggered internally at  ../torch/csrc/utils/tensor_numpy.cpp:180.)\n", "  return torch.from_numpy(parsed.astype(m[2], copy=False)).view(*s)\n", "Global seed set to 42\n"]}], "source": ["# Convert images from 0-1 to 0-255 (integers)\n", "def discretize(sample):\n", "    return (sample * 255).to(torch.int32)\n", "\n", "\n", "# Transformations applied on each image => make them a tensor and discretize\n", "transform = transforms.Compose([transforms.ToTensor(), discretize])\n", "\n", "# Loading the training dataset. We need to split it into a training and validation part\n", "train_dataset = MNIST(root=DATASET_PATH, train=True, transform=transform, download=True)\n", "pl.seed_everything(42)\n", "train_set, val_set = torch.utils.data.random_split(train_dataset, [50000, 10000])\n", "\n", "# Loading the test set\n", "test_set = MNIST(root=DATASET_PATH, train=False, transform=transform, download=True)\n", "\n", "# We define a set of data loaders that we can use for various purposes later.\n", "# Note that for actually training a model, we will use different data loaders\n", "# with a lower batch size.\n", "train_loader = data.DataLoader(train_set, batch_size=256, shuffle=False, drop_last=False)\n", "val_loader = data.DataLoader(val_set, batch_size=64, shuffle=False, drop_last=False, num_workers=4)\n", "test_loader = data.DataLoader(test_set, batch_size=64, shuffle=False, drop_last=False, num_workers=4)"]}, {"cell_type": "markdown", "id": "db5d08e8", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.010863, "end_time": "2022-05-12T12:19:08.633305", "exception": false, "start_time": "2022-05-12T12:19:08.622442", "status": "completed"}, "tags": []}, "source": ["In addition, we will define below a function to simplify the visualization of images/samples.\n", "Some training examples of the MNIST dataset is shown below."]}, {"cell_type": "code", "execution_count": 5, "id": "3c1a1c0b", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:08.657121Z", "iopub.status.busy": "2022-05-12T12:19:08.656514Z", "iopub.status.idle": "2022-05-12T12:19:08.814015Z", "shell.execute_reply": "2022-05-12T12:19:08.813306Z"}, "papermill": {"duration": 0.171707, "end_time": "2022-05-12T12:19:08.816069", "exception": false, "start_time": "2022-05-12T12:19:08.644362", "status": "completed"}, "tags": []}, "outputs": [{"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"341.674839pt\" height=\"180.72pt\" viewBox=\"0 0 341.674839 180.72\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:08.697788</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 180.72 \n", "L 341.674839 180.72 \n", "L 341.674839 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g clip-path=\"url(#p85e9a9c922)\">\n", "    <image xlink:href=\"data:image/png;base64,\n", "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\" id=\"image98de9bc152\" transform=\"scale(1 -1)translate(0 -166.32)\" x=\"7.2\" y=\"-7.2\" width=\"327.6\" height=\"166.32\"/>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"p85e9a9c922\">\n", "   <rect x=\"7.2\" y=\"7.2\" width=\"327.274839\" height=\"166.32\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 600x300 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["def show_imgs(imgs, title=None, row_size=4):\n", "    # Form a grid of pictures (we use max. 8 columns)\n", "    num_imgs = imgs.shape[0] if isinstance(imgs, Tensor) else len(imgs)\n", "    is_int = imgs.dtype == torch.int32 if isinstance(imgs, Tensor) else imgs[0].dtype == torch.int32\n", "    nrow = min(num_imgs, row_size)\n", "    ncol = int(math.ceil(num_imgs / nrow))\n", "    imgs = torchvision.utils.make_grid(imgs, nrow=nrow, pad_value=128 if is_int else 0.5)\n", "    np_imgs = imgs.cpu().numpy()\n", "    # Plot the grid\n", "    plt.figure(figsize=(1.5 * nrow, 1.5 * ncol))\n", "    plt.imshow(np.transpose(np_imgs, (1, 2, 0)), interpolation=\"nearest\")\n", "    plt.axis(\"off\")\n", "    if title is not None:\n", "        plt.title(title)\n", "    plt.show()\n", "    plt.close()\n", "\n", "\n", "show_imgs([train_set[i][0] for i in range(8)])"]}, {"cell_type": "markdown", "id": "3792e705", "metadata": {"papermill": {"duration": 0.012654, "end_time": "2022-05-12T12:19:08.842792", "exception": false, "start_time": "2022-05-12T12:19:08.830138", "status": "completed"}, "tags": []}, "source": ["## Normalizing Flows as generative model\n", "\n", "In the previous lectures, we have seen Energy-based models, Variational Autoencoders (VAEs)\n", "and Generative Adversarial Networks (GANs) as example of generative models.\n", "However, none of them explicitly learn the probability density function $p(x)$ of the real input data.\n", "While VAEs model a lower bound, energy-based models only implicitly learn the probability density.\n", "GANs on the other hand provide us a sampling mechanism for generating new data, without offering a likelihood estimate.\n", "The generative model we will look at here, called Normalizing Flows, actually models the true data distribution\n", "$p(x)$ and provides us with an exact likelihood estimate.\n", "Below, we can visually compare VAEs, GANs and Flows\n", "(figure credit - [Lilian Weng](https://lilianweng.github.io/lil-log/2018/10/13/flow-based-deep-generative-models.html)):\n", "\n", "<center width=\"100%\"><img src=\"https://github.com/PyTorchLightning/lightning-tutorials/raw/main/course_UvA-DL/09-normalizing-flows/comparison_GAN_VAE_NF.png\" width=\"600px\"></center>\n", "\n", "The major difference compared to VAEs is that flows use *invertible* functions $f$\n", "to map the input data $x$ to a latent representation $z$.\n", "To realize this, $z$ must be of the same shape as $x$.\n", "This is in contrast to VAEs where $z$ is usually much lower dimensional than the original input data.\n", "However, an invertible mapping also means that for every data point $x$, we have a corresponding latent representation\n", "$z$ which allows us to perform lossless reconstruction ($z$ to $x$).\n", "In the visualization above, this means that $x=x'$ for flows, no matter what invertible function $f$ and input $x$ we choose.\n", "\n", "Nonetheless, how are normalizing flows modeling a probability density with an invertible function?\n", "The answer to this question is the rule for change of variables.\n", "Specifically, given a prior density $p_z(z)$ (e.g. Gaussian) and an invertible function $f$,\n", "we can determine $p_x(x)$ as follows:\n", "\n", "$$\n", "\\begin{split}\n", "    \\int p_x(x) dx & = \\int p_z(z) dz = 1 \\hspace{1cm}\\text{(by definition of a probability distribution)}\\\\\n", "    \\Leftrightarrow p_x(x) & = p_z(z) \\left|\\frac{dz}{dx}\\right| = p_z(f(x)) \\left|\\frac{df(x)}{dx}\\right|\n", "\\end{split}\n", "$$\n", "\n", "Hence, in order to determine the probability of $x$, we only need to determine its probability in latent space,\n", "and get the derivate of $f$.\n", "Note that this is for a univariate distribution, and $f$ is required to be invertible and smooth.\n", "For a multivariate case, the derivative becomes a Jacobian of which we need to take the determinant.\n", "As we usually use the log-likelihood as objective, we write the multivariate term with logarithms below:\n", "\n", "$$\n", "\\log p_x(\\mathbf{x}) = \\log p_z(f(\\mathbf{x})) + \\log{} \\left|\\det \\frac{df(\\mathbf{x})}{d\\mathbf{x}}\\right|\n", "$$\n", "\n", "Although we now know how a normalizing flow obtains its likelihood, it might not be clear what a normalizing flow does intuitively.\n", "For this, we should look from the inverse perspective of the flow starting with the prior probability density $p_z(z)$.\n", "If we apply an invertible function on it, we effectively \"transform\" its probability density.\n", "For instance, if $f^{-1}(z)=z+1$, we shift the density by one while still remaining a valid probability distribution,\n", "and being invertible.\n", "We can also apply more complex transformations, like scaling: $f^{-1}(z)=2z+1$, but there you might see a difference.\n", "When you scale, you also change the volume of the probability density, as for example on uniform distributions\n", "(figure credit - [Eric Jang](https://blog.evjang.com/2018/01/nf1.html)):\n", "\n", "<center width=\"100%\"><img src=\"https://github.com/PyTorchLightning/lightning-tutorials/raw/main/course_UvA-DL/09-normalizing-flows/uniform_flow.png\" width=\"300px\"></center>\n", "\n", "You can see that the height of $p(y)$ should be lower than $p(x)$ after scaling.\n", "This change in volume represents $\\left|\\frac{df(x)}{dx}\\right|$ in our equation above,\n", "and ensures that even after scaling, we still have a valid probability distribution.\n", "We can go on with making our function $f$ more complex.\n", "However, the more complex $f$ becomes, the harder it will be to find the inverse $f^{-1}$ of it,\n", "and to calculate the log-determinant of the Jacobian $\\log{} \\left|\\det \\frac{df(\\mathbf{x})}{d\\mathbf{x}}\\right|$.\n", "An easier trick to stack multiple invertible functions $f_{1,...,K}$ after each other, as all together,\n", "they still represent a single, invertible function.\n", "Using multiple, learnable invertible functions, a normalizing flow attempts to transform\n", "$p_z(z)$ slowly into a more complex distribution which should finally be $p_x(x)$.\n", "We visualize the idea below\n", "(figure credit - [Lilian Weng](https://lilianweng.github.io/lil-log/2018/10/13/flow-based-deep-generative-models.html)):\n", "\n", "<center width=\"100%\"><img src=\"https://github.com/PyTorchLightning/lightning-tutorials/raw/main/course_UvA-DL/09-normalizing-flows/normalizing_flow_layout.png\" width=\"700px\"></center>\n", "\n", "Starting from $z_0$, which follows the prior Gaussian distribution, we sequentially apply the invertible\n", "functions $f_1,f_2,...,f_K$, until $z_K$ represents $x$.\n", "Note that in the figure above, the functions $f$ represent the inverted function from $f$ we had above\n", "(here: $f:Z\\to X$, above: $f:X\\to Z$).\n", "This is just a different notation and has no impact on the actual flow design because all $f$ need to be invertible anyways.\n", "When we estimate the log likelihood of a data point $x$ as in the equations above,\n", "we run the flows in the opposite direction than visualized above.\n", "Multiple flow layers have been proposed that use a neural network as learnable parameters,\n", "such as the planar and radial flow.\n", "However, we will focus here on flows that are commonly used in image\n", "modeling, and will discuss them in the rest of the notebook along with\n", "the details of how to train a normalizing flow."]}, {"cell_type": "markdown", "id": "a5b52653", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.012573, "end_time": "2022-05-12T12:19:08.868607", "exception": false, "start_time": "2022-05-12T12:19:08.856034", "status": "completed"}, "tags": []}, "source": ["## Normalizing Flows on images\n", "\n", "<div class=\"center-wrapper\"><div class=\"video-wrapper\"><iframe src=\"https://www.youtube.com/embed/qMoGcRhVrF8\" title=\"YouTube video player\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture\" allowfullscreen></iframe></div></div>\n", "\n", "To become familiar with normalizing flows, especially for the application of image modeling,\n", "it is best to discuss the different elements in a flow along with the implementation.\n", "As a general concept, we want to build a normalizing flow that maps an input image (here MNIST) to an equally sized latent space:\n", "\n", "<center width=\"100%\" style=\"padding: 10px\"><img src=\"https://github.com/PyTorchLightning/lightning-tutorials/raw/main/course_UvA-DL/09-normalizing-flows/image_to_gaussian.svg\" width=\"450px\"></center>\n", "\n", "As a first step, we will implement a template of a normalizing flow in PyTorch Lightning.\n", "During training and validation, a normalizing flow performs density estimation in the forward direction.\n", "For this, we apply a series of flow transformations on the input $x$ and estimate the probability\n", "of the input by determining the probability of the transformed point $z$ given a prior,\n", "and the change of volume caused by the transformations.\n", "During inference, we can do both density estimation and sampling new points by inverting the flow transformations.\n", "Therefore, we define a function `_get_likelihood` which performs density estimation,\n", "and `sample` to generate new examples.\n", "The functions `training_step`, `validation_step` and `test_step` all make use of `_get_likelihood`.\n", "\n", "The standard metric used in generative models, and in particular normalizing flows, is bits per dimensions (bpd).\n", "Bpd is motivated from an information theory perspective and describes how many bits we would need to encode a particular example in our modeled distribution.\n", "The less bits we need, the more likely the example in our distribution.\n", "When we test for the bits per dimension of our test dataset, we can judge whether our model generalizes to new samples of the dataset and didn't memorize the training dataset.\n", "In order to calculate the bits per dimension score, we can rely on the negative log-likelihood and change the log base (as bits are binary while NLL is usually exponential):\n", "\n", "$$\\text{bpd} = \\text{nll} \\cdot \\log_2\\left(\\exp(1)\\right) \\cdot \\left(\\prod d_i\\right)^{-1}$$\n", "\n", "where $d_1,...,d_K$ are the dimensions of the input.\n", "For images, this would be the height, width and channel number.\n", "We divide the log likelihood by these extra dimensions to have a metric which we can compare for different image resolutions.\n", "In the original image space, MNIST examples have a bits per dimension\n", "score of 8 (we need 8 bits to encode each pixel as there are 256\n", "possible values)."]}, {"cell_type": "code", "execution_count": 6, "id": "9d6d5907", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:08.893950Z", "iopub.status.busy": "2022-05-12T12:19:08.893239Z", "iopub.status.idle": "2022-05-12T12:19:08.906436Z", "shell.execute_reply": "2022-05-12T12:19:08.905728Z"}, "lines_to_next_cell": 2, "papermill": {"duration": 0.026966, "end_time": "2022-05-12T12:19:08.907991", "exception": false, "start_time": "2022-05-12T12:19:08.881025", "status": "completed"}, "tags": []}, "outputs": [], "source": ["class ImageFlow(pl.LightningModule):\n", "    def __init__(self, flows, import_samples=8):\n", "        \"\"\"\n", "        Args:\n", "            flows: A list of flows (each a nn.Module) that should be applied on the images.\n", "            import_samples: Number of importance samples to use during testing (see explanation below). Can be changed at any time\n", "        \"\"\"\n", "        super().__init__()\n", "        self.flows = nn.ModuleList(flows)\n", "        self.import_samples = import_samples\n", "        # Create prior distribution for final latent space\n", "        self.prior = torch.distributions.normal.Normal(loc=0.0, scale=1.0)\n", "        # Example input for visualizing the graph\n", "        self.example_input_array = train_set[0][0].unsqueeze(dim=0)\n", "\n", "    def forward(self, imgs):\n", "        # The forward function is only used for visualizing the graph\n", "        return self._get_likelihood(imgs)\n", "\n", "    def encode(self, imgs):\n", "        # Given a batch of images, return the latent representation z and ldj of the transformations\n", "        z, ldj = imgs, torch.zeros(imgs.shape[0], device=self.device)\n", "        for flow in self.flows:\n", "            z, ldj = flow(z, ldj, reverse=False)\n", "        return z, ldj\n", "\n", "    def _get_likelihood(self, imgs, return_ll=False):\n", "        \"\"\"Given a batch of images, return the likelihood of those.\n", "\n", "        If return_ll is True, this function returns the log likelihood of the input. Otherwise, the ouptut metric is\n", "        bits per dimension (scaled negative log likelihood)\n", "        \"\"\"\n", "        z, ldj = self.encode(imgs)\n", "        log_pz = self.prior.log_prob(z).sum(dim=[1, 2, 3])\n", "        log_px = ldj + log_pz\n", "        nll = -log_px\n", "        # Calculating bits per dimension\n", "        bpd = nll * np.log2(np.exp(1)) / np.prod(imgs.shape[1:])\n", "        return bpd.mean() if not return_ll else log_px\n", "\n", "    @torch.no_grad()\n", "    def sample(self, img_shape, z_init=None):\n", "        \"\"\"Sample a batch of images from the flow.\"\"\"\n", "        # Sample latent representation from prior\n", "        if z_init is None:\n", "            z = self.prior.sample(sample_shape=img_shape).to(device)\n", "        else:\n", "            z = z_init.to(device)\n", "\n", "        # Transform z to x by inverting the flows\n", "        ldj = torch.zeros(img_shape[0], device=device)\n", "        for flow in reversed(self.flows):\n", "            z, ldj = flow(z, ldj, reverse=True)\n", "        return z\n", "\n", "    def configure_optimizers(self):\n", "        optimizer = optim.Adam(self.parameters(), lr=1e-3)\n", "        # An scheduler is optional, but can help in flows to get the last bpd improvement\n", "        scheduler = optim.lr_scheduler.StepLR(optimizer, 1, gamma=0.99)\n", "        return [optimizer], [scheduler]\n", "\n", "    def training_step(self, batch, batch_idx):\n", "        # Normalizing flows are trained by maximum likelihood => return bpd\n", "        loss = self._get_likelihood(batch[0])\n", "        self.log(\"train_bpd\", loss)\n", "        return loss\n", "\n", "    def validation_step(self, batch, batch_idx):\n", "        loss = self._get_likelihood(batch[0])\n", "        self.log(\"val_bpd\", loss)\n", "\n", "    def test_step(self, batch, batch_idx):\n", "        # Perform importance sampling during testing => estimate likelihood M times for each image\n", "        samples = []\n", "        for _ in range(self.import_samples):\n", "            img_ll = self._get_likelihood(batch[0], return_ll=True)\n", "            samples.append(img_ll)\n", "        img_ll = torch.stack(samples, dim=-1)\n", "\n", "        # To average the probabilities, we need to go from log-space to exp, and back to log.\n", "        # Logsumexp provides us a stable implementation for this\n", "        img_ll = torch.logsumexp(img_ll, dim=-1) - np.log(self.import_samples)\n", "\n", "        # Calculate final bpd\n", "        bpd = -img_ll * np.log2(np.exp(1)) / np.prod(batch[0].shape[1:])\n", "        bpd = bpd.mean()\n", "\n", "        self.log(\"test_bpd\", bpd)"]}, {"cell_type": "markdown", "id": "98940a88", "metadata": {"papermill": {"duration": 0.010911, "end_time": "2022-05-12T12:19:08.929952", "exception": false, "start_time": "2022-05-12T12:19:08.919041", "status": "completed"}, "tags": []}, "source": ["The `test_step` function differs from the training and validation step in that it makes use of importance sampling.\n", "We will discuss the motiviation and details behind this after\n", "understanding how flows model discrete images in continuous space."]}, {"cell_type": "markdown", "id": "32009fa8", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.011237, "end_time": "2022-05-12T12:19:08.952316", "exception": false, "start_time": "2022-05-12T12:19:08.941079", "status": "completed"}, "tags": []}, "source": ["### Dequantization\n", "\n", "Normalizing flows rely on the rule of change of variables, which is naturally defined in continuous space.\n", "Applying flows directly on discrete data leads to undesired density models where arbitrarly high likelihood are placed on a few, particular values.\n", "See the illustration below:\n", "\n", "<center><img src=\"https://github.com/PyTorchLightning/lightning-tutorials/raw/main/course_UvA-DL/09-normalizing-flows/dequantization_issue.svg\" width=\"40%\"/></center>\n", "\n", "The black points represent the discrete points, and the green volume the density modeled by a normalizing flow in continuous space.\n", "The flow would continue to increase the likelihood for $x=0,1,2,3$ while having no volume on any other point.\n", "Remember that in continuous space, we have the constraint that the overall volume of the probability density must be 1 ($\\int p(x)dx=1$).\n", "Otherwise, we don't model a probability distribution anymore.\n", "However, the discrete points $x=0,1,2,3$ represent delta peaks with no width in continuous space.\n", "This is why the flow can place an infinite high likelihood on these few points while still representing a distribution in continuous space.\n", "Nonetheless, the learned density does not tell us anything about the distribution among the discrete points,\n", "as in discrete space, the likelihoods of those four points would have to sum to 1, not to infinity.\n", "\n", "To prevent such degenerated solutions, a common solution is to add a small amount of noise to each discrete value, which is also referred to as dequantization.\n", "Considering $x$ as an integer (as it is the case for images), the dequantized representation $v$ can be formulated as $v=x+u$ where $u\\in[0,1)^D$.\n", "Thus, the discrete value $1$ is modeled by a distribution over the interval $[1.0, 2.0)$, the value $2$ by an volume over $[2.0, 3.0)$, etc.\n", "Our objective of modeling $p(x)$ becomes:\n", "\n", "$$ p(x) = \\int p(x+u)du = \\int \\frac{q(u|x)}{q(u|x)}p(x+u)du = \\mathbb{E}_{u\\sim q(u|x)}\\left[\\frac{p(x+u)}{q(u|x)} \\right]$$\n", "\n", "with $q(u|x)$ being the noise distribution.\n", "For now, we assume it to be uniform, which can also be written as $p(x)=\\mathbb{E}_{u\\sim U(0,1)^D}\\left[p(x+u) \\right]$.\n", "\n", "In the following, we will implement Dequantization as a flow transformation itself.\n", "After adding noise to the discrete values, we additionally transform the volume into a Gaussian-like shape.\n", "This is done by scaling $x+u$ between $0$ and $1$, and applying the invert of the sigmoid function $\\sigma(z)^{-1} = \\log z - \\log 1-z$.\n", "If we would not do this, we would face two problems:\n", "\n", "1.\n", "The input is scaled between 0 and 256 while the prior distribution is a Gaussian with mean $0$ and standard deviation $1$.\n", "In the first iterations after initializing the parameters of the flow, we would have extremely low likelihoods for large values like $256$.\n", "This would cause the training to diverge instantaneously.\n", "2.\n", "As the output distribution is a Gaussian, it is beneficial for the flow to have a similarly shaped input distribution.\n", "This will reduce the modeling complexity that is required by the flow.\n", "\n", "Overall, we can implement dequantization as follows:"]}, {"cell_type": "code", "execution_count": 7, "id": "f9117a9f", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:08.976090Z", "iopub.status.busy": "2022-05-12T12:19:08.975369Z", "iopub.status.idle": "2022-05-12T12:19:08.984311Z", "shell.execute_reply": "2022-05-12T12:19:08.983622Z"}, "papermill": {"duration": 0.022493, "end_time": "2022-05-12T12:19:08.985814", "exception": false, "start_time": "2022-05-12T12:19:08.963321", "status": "completed"}, "tags": []}, "outputs": [], "source": ["class Dequantization(nn.Module):\n", "    def __init__(self, alpha=1e-5, quants=256):\n", "        \"\"\"\n", "        Args:\n", "            alpha: small constant that is used to scale the original input.\n", "                    Prevents dealing with values very close to 0 and 1 when inverting the sigmoid\n", "            quants: Number of possible discrete values (usually 256 for 8-bit image)\n", "        \"\"\"\n", "        super().__init__()\n", "        self.alpha = alpha\n", "        self.quants = quants\n", "\n", "    def forward(self, z, ldj, reverse=False):\n", "        if not reverse:\n", "            z, ldj = self.dequant(z, ldj)\n", "            z, ldj = self.sigmoid(z, ldj, reverse=True)\n", "        else:\n", "            z, ldj = self.sigmoid(z, ldj, reverse=False)\n", "            z = z * self.quants\n", "            ldj += np.log(self.quants) * np.prod(z.shape[1:])\n", "            z = torch.floor(z).clamp(min=0, max=self.quants - 1).to(torch.int32)\n", "        return z, ldj\n", "\n", "    def sigmoid(self, z, ldj, reverse=False):\n", "        # Applies an invertible sigmoid transformation\n", "        if not reverse:\n", "            ldj += (-z - 2 * F.softplus(-z)).sum(dim=[1, 2, 3])\n", "            z = torch.sigmoid(z)\n", "        else:\n", "            z = z * (1 - self.alpha) + 0.5 * self.alpha  # Scale to prevent boundaries 0 and 1\n", "            ldj += np.log(1 - self.alpha) * np.prod(z.shape[1:])\n", "            ldj += (-torch.log(z) - torch.log(1 - z)).sum(dim=[1, 2, 3])\n", "            z = torch.log(z) - torch.log(1 - z)\n", "        return z, ldj\n", "\n", "    def dequant(self, z, ldj):\n", "        # Transform discrete values to continuous volumes\n", "        z = z.to(torch.float32)\n", "        z = z + torch.rand_like(z).detach()\n", "        z = z / self.quants\n", "        ldj -= np.log(self.quants) * np.prod(z.shape[1:])\n", "        return z, ldj"]}, {"cell_type": "markdown", "id": "61405f9c", "metadata": {"papermill": {"duration": 0.01093, "end_time": "2022-05-12T12:19:09.009683", "exception": false, "start_time": "2022-05-12T12:19:08.998753", "status": "completed"}, "tags": []}, "source": ["A good check whether a flow is correctly implemented or not, is to verify that it is invertible.\n", "Hence, we will dequantize a randomly chosen training image, and then quantize it again.\n", "We would expect that we would get the exact same image out:"]}, {"cell_type": "code", "execution_count": 8, "id": "6e314bbd", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:09.032925Z", "iopub.status.busy": "2022-05-12T12:19:09.032430Z", "iopub.status.idle": "2022-05-12T12:19:09.042478Z", "shell.execute_reply": "2022-05-12T12:19:09.041770Z"}, "papermill": {"duration": 0.023233, "end_time": "2022-05-12T12:19:09.043890", "exception": false, "start_time": "2022-05-12T12:19:09.020657", "status": "completed"}, "tags": []}, "outputs": [{"name": "stderr", "output_type": "stream", "text": ["Global seed set to 42\n"]}, {"name": "stdout", "output_type": "stream", "text": ["Dequantization was not invertible.\n", "Original value: 0\n", "Reconstructed value: 1\n"]}], "source": ["# Testing invertibility of dequantization layer\n", "pl.seed_everything(42)\n", "orig_img = train_set[0][0].unsqueeze(dim=0)\n", "ldj = torch.zeros(\n", "    1,\n", ")\n", "dequant_module = Dequantization()\n", "deq_img, ldj = dequant_module(orig_img, ldj, reverse=False)\n", "reconst_img, ldj = dequant_module(deq_img, ldj, reverse=True)\n", "\n", "d1, d2 = torch.where(orig_img.squeeze() != reconst_img.squeeze())\n", "if len(d1) != 0:\n", "    print(\"Dequantization was not invertible.\")\n", "    for i in range(d1.shape[0]):\n", "        print(\"Original value:\", orig_img[0, 0, d1[i], d2[i]].item())\n", "        print(\"Reconstructed value:\", reconst_img[0, 0, d1[i], d2[i]].item())\n", "else:\n", "    print(\"Successfully inverted dequantization\")\n", "\n", "# Layer is not strictly invertible due to float precision constraints\n", "# assert (orig_img == reconst_img).all().item()"]}, {"cell_type": "markdown", "id": "e4bb61fc", "metadata": {"papermill": {"duration": 0.011138, "end_time": "2022-05-12T12:19:09.066797", "exception": false, "start_time": "2022-05-12T12:19:09.055659", "status": "completed"}, "tags": []}, "source": ["In contrast to our expectation, the test fails.\n", "However, this is no reason to doubt our implementation here as only one single value is not equal to the original.\n", "This is caused due to numerical inaccuracies in the sigmoid invert.\n", "While the input space to the inverted sigmoid is scaled between 0 and 1, the output space is between $-\\infty$ and $\\infty$.\n", "And as we use 32 bits to represent the numbers (in addition to applying logs over and over again),\n", "such inaccuries can occur and should not be worrisome.\n", "Nevertheless, it is good to be aware of them, and can be improved by using a double tensor (float64).\n", "\n", "Finally, we can take our dequantization and actually visualize the\n", "distribution it transforms the discrete values into:"]}, {"cell_type": "code", "execution_count": 9, "id": "51af4cde", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:09.091044Z", "iopub.status.busy": "2022-05-12T12:19:09.090281Z", "iopub.status.idle": "2022-05-12T12:19:09.776862Z", "shell.execute_reply": "2022-05-12T12:19:09.776116Z"}, "papermill": {"duration": 0.701406, "end_time": "2022-05-12T12:19:09.779430", "exception": false, "start_time": "2022-05-12T12:19:09.078024", "status": "completed"}, "tags": []}, "outputs": [{"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"393.732813pt\" height=\"226.194375pt\" viewBox=\"0 0 393.732813 226.194375\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:09.492340</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 226.194375 \n", "L 393.732813 226.194375 \n", "L 393.732813 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g id=\"patch_2\">\n", "    <path d=\"M 43.78125 188.638125 \n", "L 378.58125 188.638125 \n", "L 378.58125 22.318125 \n", "L 43.78125 22.318125 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "   </g>\n", "   <g id=\"matplotlib.axis_1\">\n", "    <g id=\"xtick_1\">\n", "     <g id=\"text_1\">\n", "      <!-- -4.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(34.025781 203.236563)scale(0.1 -0.1)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-2d\" d=\"M 313 2009 \n", "L 1997 2009 \n", "L 1997 1497 \n", "L 313 1497 \n", "L 313 2009 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-34\" d=\"M 2419 4116 \n", "L 825 1625 \n", "L 2419 1625 \n", "L 2419 4116 \n", "z\n", "M 2253 4666 \n", "L 3047 4666 \n", "L 3047 1625 \n", "L 3713 1625 \n", "L 3713 1100 \n", "L 3047 1100 \n", "L 3047 0 \n", "L 2419 0 \n", "L 2419 1100 \n", "L 313 1100 \n", "L 313 1709 \n", "L 2253 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-2e\" d=\"M 684 794 \n", "L 1344 794 \n", "L 1344 0 \n", "L 684 0 \n", "L 684 794 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-30\" d=\"M 2034 4250 \n", "Q 1547 4250 1301 3770 \n", "Q 1056 3291 1056 2328 \n", "Q 1056 1369 1301 889 \n", "Q 1547 409 2034 409 \n", "Q 2525 409 2770 889 \n", "Q 3016 1369 3016 2328 \n", "Q 3016 3291 2770 3770 \n", "Q 2525 4250 2034 4250 \n", "z\n", "M 2034 4750 \n", "Q 2819 4750 3233 4129 \n", "Q 3647 3509 3647 2328 \n", "Q 3647 1150 3233 529 \n", "Q 2819 -91 2034 -91 \n", "Q 1250 -91 836 529 \n", "Q 422 1150 422 2328 \n", "Q 422 3509 836 4129 \n", "Q 1250 4750 2034 4750 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2d\"/>\n", "       <use xlink:href=\"#DejaVuSans-34\" x=\"36.083984\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"99.707031\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"131.494141\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_2\">\n", "     <g id=\"text_2\">\n", "      <!-- -1.9 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(120.344677 203.236563)scale(0.1 -0.1)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-31\" d=\"M 794 531 \n", "L 1825 531 \n", "L 1825 4091 \n", "L 703 3866 \n", "L 703 4441 \n", "L 1819 4666 \n", "L 2450 4666 \n", "L 2450 531 \n", "L 3481 531 \n", "L 3481 0 \n", "L 794 0 \n", "L 794 531 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-39\" d=\"M 703 97 \n", "L 703 672 \n", "Q 941 559 1184 500 \n", "Q 1428 441 1663 441 \n", "Q 2288 441 2617 861 \n", "Q 2947 1281 2994 2138 \n", "Q 2813 1869 2534 1725 \n", "Q 2256 1581 1919 1581 \n", "Q 1219 1581 811 2004 \n", "Q 403 2428 403 3163 \n", "Q 403 3881 828 4315 \n", "Q 1253 4750 1959 4750 \n", "Q 2769 4750 3195 4129 \n", "Q 3622 3509 3622 2328 \n", "Q 3622 1225 3098 567 \n", "Q 2575 -91 1691 -91 \n", "Q 1453 -91 1209 -44 \n", "Q 966 3 703 97 \n", "z\n", "M 1959 2075 \n", "Q 2384 2075 2632 2365 \n", "Q 2881 2656 2881 3163 \n", "Q 2881 3666 2632 3958 \n", "Q 2384 4250 1959 4250 \n", "Q 1534 4250 1286 3958 \n", "Q 1038 3666 1038 3163 \n", "Q 1038 2656 1286 2365 \n", "Q 1534 2075 1959 2075 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2d\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"36.083984\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"99.707031\"/>\n", "       <use xlink:href=\"#DejaVuSans-39\" x=\"131.494141\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_3\">\n", "     <g id=\"text_3\">\n", "      <!-- -1.1 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(155.9617 203.236563)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-2d\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"36.083984\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"99.707031\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"131.494141\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_4\">\n", "     <g id=\"text_4\">\n", "      <!-- -0.5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(180.265081 203.236563)scale(0.1 -0.1)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-35\" d=\"M 691 4666 \n", "L 3169 4666 \n", "L 3169 4134 \n", "L 1269 4134 \n", "L 1269 2991 \n", "Q 1406 3038 1543 3061 \n", "Q 1681 3084 1819 3084 \n", "Q 2600 3084 3056 2656 \n", "Q 3513 2228 3513 1497 \n", "Q 3513 744 3044 326 \n", "Q 2575 -91 1722 -91 \n", "Q 1428 -91 1123 -41 \n", "Q 819 9 494 109 \n", "L 494 744 \n", "Q 775 591 1075 516 \n", "Q 1375 441 1709 441 \n", "Q 2250 441 2565 725 \n", "Q 2881 1009 2881 1497 \n", "Q 2881 1984 2565 2268 \n", "Q 2250 2553 1709 2553 \n", "Q 1456 2553 1204 2497 \n", "Q 953 2441 691 2322 \n", "L 691 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2d\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"36.083984\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"99.707031\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"131.494141\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_5\">\n", "     <g id=\"text_5\">\n", "      <!-- 0.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(203.439199 203.236563)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_6\">\n", "     <g id=\"text_6\">\n", "      <!-- 0.5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(225.228435 203.236563)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_7\">\n", "     <g id=\"text_7\">\n", "      <!-- 1.1 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(249.531811 203.236563)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-31\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_8\">\n", "     <g id=\"text_8\">\n", "      <!-- 1.9 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(285.148833 203.236563)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-31\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-39\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_9\">\n", "     <g id=\"text_9\">\n", "      <!-- 4.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(370.629688 203.236563)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-34\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"text_10\">\n", "     <!-- z -->\n", "     <g style=\"fill: #262626\" transform=\"translate(208.557031 216.914688)scale(0.1 -0.1)\">\n", "      <defs>\n", "       <path id=\"DejaVuSans-7a\" d=\"M 353 3500 \n", "L 3084 3500 \n", "L 3084 2975 \n", "L 922 459 \n", "L 3084 459 \n", "L 3084 0 \n", "L 275 0 \n", "L 275 525 \n", "L 2438 3041 \n", "L 353 3041 \n", "L 353 3500 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      </defs>\n", "      <use xlink:href=\"#DejaVuSans-7a\"/>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"matplotlib.axis_2\">\n", "    <g id=\"ytick_1\">\n", "     <g id=\"text_11\">\n", "      <!-- 0.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(20.878125 192.437344)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_2\">\n", "     <g id=\"text_12\">\n", "      <!-- 0.5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(20.878125 154.637344)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_3\">\n", "     <g id=\"text_13\">\n", "      <!-- 1.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(20.878125 116.837344)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-31\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_4\">\n", "     <g id=\"text_14\">\n", "      <!-- 1.5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(20.878125 79.037344)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-31\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_5\">\n", "     <g id=\"text_15\">\n", "      <!-- 2.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(20.878125 41.237344)scale(0.1 -0.1)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-32\" d=\"M 1228 531 \n", "L 3431 531 \n", "L 3431 0 \n", "L 469 0 \n", "L 469 531 \n", "Q 828 903 1448 1529 \n", "Q 2069 2156 2228 2338 \n", "Q 2531 2678 2651 2914 \n", "Q 2772 3150 2772 3378 \n", "Q 2772 3750 2511 3984 \n", "Q 2250 4219 1831 4219 \n", "Q 1534 4219 1204 4116 \n", "Q 875 4013 500 3803 \n", "L 500 4441 \n", "Q 881 4594 1212 4672 \n", "Q 1544 4750 1819 4750 \n", "Q 2544 4750 2975 4387 \n", "Q 3406 4025 3406 3419 \n", "Q 3406 3131 3298 2873 \n", "Q 3191 2616 2906 2266 \n", "Q 2828 2175 2409 1742 \n", "Q 1991 1309 1228 531 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-32\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"text_16\">\n", "     <!-- Probability -->\n", "     <g style=\"fill: #262626\" transform=\"translate(14.798438 131.907031)rotate(-90)scale(0.1 -0.1)\">\n", "      <defs>\n", "       <path id=\"DejaVuSans-50\" d=\"M 1259 4147 \n", "L 1259 2394 \n", "L 2053 2394 \n", "Q 2494 2394 2734 2622 \n", "Q 2975 2850 2975 3272 \n", "Q 2975 3691 2734 3919 \n", "Q 2494 4147 2053 4147 \n", "L 1259 4147 \n", "z\n", "M 628 4666 \n", "L 2053 4666 \n", "Q 2838 4666 3239 4311 \n", "Q 3641 3956 3641 3272 \n", "Q 3641 2581 3239 2228 \n", "Q 2838 1875 2053 1875 \n", "L 1259 1875 \n", "L 1259 0 \n", "L 628 0 \n", "L 628 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-72\" d=\"M 2631 2963 \n", "Q 2534 3019 2420 3045 \n", "Q 2306 3072 2169 3072 \n", "Q 1681 3072 1420 2755 \n", "Q 1159 2438 1159 1844 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 3500 \n", "L 1159 3500 \n", "L 1159 2956 \n", "Q 1341 3275 1631 3429 \n", "Q 1922 3584 2338 3584 \n", "Q 2397 3584 2469 3576 \n", "Q 2541 3569 2628 3553 \n", "L 2631 2963 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-6f\" d=\"M 1959 3097 \n", "Q 1497 3097 1228 2736 \n", "Q 959 2375 959 1747 \n", "Q 959 1119 1226 758 \n", "Q 1494 397 1959 397 \n", "Q 2419 397 2687 759 \n", "Q 2956 1122 2956 1747 \n", "Q 2956 2369 2687 2733 \n", "Q 2419 3097 1959 3097 \n", "z\n", "M 1959 3584 \n", "Q 2709 3584 3137 3096 \n", "Q 3566 2609 3566 1747 \n", "Q 3566 888 3137 398 \n", "Q 2709 -91 1959 -91 \n", "Q 1206 -91 779 398 \n", "Q 353 888 353 1747 \n", "Q 353 2609 779 3096 \n", "Q 1206 3584 1959 3584 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-62\" d=\"M 3116 1747 \n", "Q 3116 2381 2855 2742 \n", "Q 2594 3103 2138 3103 \n", "Q 1681 3103 1420 2742 \n", "Q 1159 2381 1159 1747 \n", "Q 1159 1113 1420 752 \n", "Q 1681 391 2138 391 \n", "Q 2594 391 2855 752 \n", "Q 3116 1113 3116 1747 \n", "z\n", "M 1159 2969 \n", "Q 1341 3281 1617 3432 \n", "Q 1894 3584 2278 3584 \n", "Q 2916 3584 3314 3078 \n", "Q 3713 2572 3713 1747 \n", "Q 3713 922 3314 415 \n", "Q 2916 -91 2278 -91 \n", "Q 1894 -91 1617 61 \n", "Q 1341 213 1159 525 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 4863 \n", "L 1159 4863 \n", "L 1159 2969 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-61\" d=\"M 2194 1759 \n", "Q 1497 1759 1228 1600 \n", "Q 959 1441 959 1056 \n", "Q 959 750 1161 570 \n", "Q 1363 391 1709 391 \n", "Q 2188 391 2477 730 \n", "Q 2766 1069 2766 1631 \n", "L 2766 1759 \n", "L 2194 1759 \n", "z\n", "M 3341 1997 \n", "L 3341 0 \n", "L 2766 0 \n", "L 2766 531 \n", "Q 2569 213 2275 61 \n", "Q 1981 -91 1556 -91 \n", "Q 1019 -91 701 211 \n", "Q 384 513 384 1019 \n", "Q 384 1609 779 1909 \n", "Q 1175 2209 1959 2209 \n", "L 2766 2209 \n", "L 2766 2266 \n", "Q 2766 2663 2505 2880 \n", "Q 2244 3097 1772 3097 \n", "Q 1472 3097 1187 3025 \n", "Q 903 2953 641 2809 \n", "L 641 3341 \n", "Q 956 3463 1253 3523 \n", "Q 1550 3584 1831 3584 \n", "Q 2591 3584 2966 3190 \n", "Q 3341 2797 3341 1997 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-69\" d=\"M 603 3500 \n", "L 1178 3500 \n", "L 1178 0 \n", "L 603 0 \n", "L 603 3500 \n", "z\n", "M 603 4863 \n", "L 1178 4863 \n", "L 1178 4134 \n", "L 603 4134 \n", "L 603 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-6c\" d=\"M 603 4863 \n", "L 1178 4863 \n", "L 1178 0 \n", "L 603 0 \n", "L 603 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-74\" d=\"M 1172 4494 \n", "L 1172 3500 \n", "L 2356 3500 \n", "L 2356 3053 \n", "L 1172 3053 \n", "L 1172 1153 \n", "Q 1172 725 1289 603 \n", "Q 1406 481 1766 481 \n", "L 2356 481 \n", "L 2356 0 \n", "L 1766 0 \n", "Q 1100 0 847 248 \n", "Q 594 497 594 1153 \n", "L 594 3053 \n", "L 172 3053 \n", "L 172 3500 \n", "L 594 3500 \n", "L 594 4494 \n", "L 1172 4494 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-79\" d=\"M 2059 -325 \n", "Q 1816 -950 1584 -1140 \n", "Q 1353 -1331 966 -1331 \n", "L 506 -1331 \n", "L 506 -850 \n", "L 844 -850 \n", "Q 1081 -850 1212 -737 \n", "Q 1344 -625 1503 -206 \n", "L 1606 56 \n", "L 191 3500 \n", "L 800 3500 \n", "L 1894 763 \n", "L 2988 3500 \n", "L 3597 3500 \n", "L 2059 -325 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      </defs>\n", "      <use xlink:href=\"#DejaVuSans-50\"/>\n", "      <use xlink:href=\"#DejaVuSans-72\" x=\"58.552734\"/>\n", "      <use xlink:href=\"#DejaVuSans-6f\" x=\"97.416016\"/>\n", "      <use xlink:href=\"#DejaVuSans-62\" x=\"158.597656\"/>\n", "      <use xlink:href=\"#DejaVuSans-61\" x=\"222.074219\"/>\n", "      <use xlink:href=\"#DejaVuSans-62\" x=\"283.353516\"/>\n", "      <use xlink:href=\"#DejaVuSans-69\" x=\"346.830078\"/>\n", "      <use xlink:href=\"#DejaVuSans-6c\" x=\"374.613281\"/>\n", "      <use xlink:href=\"#DejaVuSans-69\" x=\"402.396484\"/>\n", "      <use xlink:href=\"#DejaVuSans-74\" x=\"430.179688\"/>\n", "      <use xlink:href=\"#DejaVuSans-79\" x=\"469.388672\"/>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_1\">\n", "    <defs>\n", "     <path id=\"mbc49098ec0\" d=\"M 43.78125 -37.55625 \n", "L 43.78125 -48.238657 \n", "L 44.200273 -48.342109 \n", "L 44.619297 -48.446538 \n", "L 45.03832 -48.551936 \n", "L 45.457344 -48.658306 \n", "L 45.876367 -48.765666 \n", "L 46.29539 -48.874031 \n", "L 46.714414 -48.98339 \n", "L 47.133437 -49.093766 \n", "L 47.55246 -49.205165 \n", "L 47.971484 -49.317593 \n", "L 48.390507 -49.431062 \n", "L 48.809531 -49.545573 \n", "L 49.228554 -49.661149 \n", "L 49.647577 -49.777787 \n", "L 50.066601 -49.895495 \n", "L 50.485634 -50.014286 \n", "L 50.904657 -50.134177 \n", "L 51.323681 -50.255161 \n", "L 51.742704 -50.377258 \n", "L 52.161728 -50.500469 \n", "L 52.580751 -50.624814 \n", "L 52.999774 -50.750297 \n", "L 53.418798 -50.876921 \n", "L 53.837821 -51.004708 \n", "L 54.256844 -51.133657 \n", "L 54.675868 -51.263777 \n", "L 55.094891 -51.395085 \n", "L 55.513915 -51.527584 \n", "L 55.932938 -51.66129 \n", "L 56.351961 -51.796209 \n", "L 56.770985 -51.93235 \n", "L 57.190008 -52.069722 \n", "L 57.609032 -52.208345 \n", "L 58.028055 -52.348206 \n", "L 58.447078 -52.48934 \n", "L 58.866102 -52.631741 \n", "L 59.285125 -52.775426 \n", "L 59.704148 -52.920401 \n", "L 60.123172 -53.066683 \n", "L 60.542205 -53.21428 \n", "L 60.961229 -53.363193 \n", "L 61.380252 -53.513441 \n", "L 61.799275 -53.665031 \n", "L 62.218299 -53.817968 \n", "L 62.637322 -53.972279 \n", "L 63.056345 -54.127959 \n", "L 63.475369 -54.285028 \n", "L 63.894392 -54.44349 \n", "L 64.313416 -54.603354 \n", "L 64.732439 -54.764641 \n", "L 65.151462 -54.927347 \n", "L 65.570486 -55.0915 \n", "L 65.989509 -55.257094 \n", "L 66.408532 -55.424149 \n", "L 66.827556 -55.592677 \n", "L 67.246579 -55.762684 \n", "L 67.665603 -55.934187 \n", "L 68.084626 -56.107181 \n", "L 68.503649 -56.281697 \n", "L 68.922673 -56.457737 \n", "L 69.341696 -56.635314 \n", "L 69.76072 -56.814441 \n", "L 70.179743 -56.995118 \n", "L 70.598776 -57.177374 \n", "L 71.0178 -57.361203 \n", "L 71.436823 -57.546628 \n", "L 71.855846 -57.733647 \n", "L 72.27487 -57.922288 \n", "L 72.693893 -58.112546 \n", "L 73.112917 -58.304443 \n", "L 73.53194 -58.497991 \n", "L 73.950963 -58.693205 \n", "L 74.369987 -58.890075 \n", "L 74.78901 -59.088635 \n", "L 75.208033 -59.288886 \n", "L 75.627057 -59.490845 \n", "L 76.04608 -59.69451 \n", "L 76.465104 -59.899908 \n", "L 76.884127 -60.107049 \n", "L 77.30315 -60.315948 \n", "L 77.722174 -60.526595 \n", "L 78.141197 -60.739016 \n", "L 78.56022 -60.953234 \n", "L 78.979244 -61.169239 \n", "L 79.398267 -61.387058 \n", "L 79.817291 -61.60669 \n", "L 80.236314 -61.82815 \n", "L 80.655347 -62.051468 \n", "L 81.074371 -62.276634 \n", "L 81.493394 -62.503656 \n", "L 81.912418 -62.732571 \n", "L 82.331441 -62.963367 \n", "L 82.750464 -63.196061 \n", "L 83.169488 -63.430662 \n", "L 83.588511 -63.667199 \n", "L 84.007534 -63.905661 \n", "L 84.426558 -64.146071 \n", "L 84.845581 -64.388443 \n", "L 85.264605 -64.63278 \n", "L 85.683628 -64.879095 \n", "L 86.102651 -65.1274 \n", "L 86.521675 -65.377712 \n", "L 86.940698 -65.630039 \n", "L 87.359721 -65.884386 \n", "L 87.778745 -66.140777 \n", "L 88.197768 -66.399216 \n", "L 88.616792 -66.659712 \n", "L 89.035815 -66.922281 \n", "L 89.454838 -67.186936 \n", "L 89.873862 -67.453678 \n", "L 90.292885 -67.722517 \n", "L 90.711908 -67.993472 \n", "L 91.130932 -68.266551 \n", "L 91.549955 -68.541769 \n", "L 91.968979 -68.819126 \n", "L 92.388002 -69.098652 \n", "L 92.807025 -69.380327 \n", "L 93.226049 -69.664193 \n", "L 93.645072 -69.950255 \n", "L 94.064106 -70.238497 \n", "L 94.483129 -70.528968 \n", "L 94.902152 -70.821638 \n", "L 95.321176 -71.116538 \n", "L 95.740199 -71.413682 \n", "L 96.159222 -71.713082 \n", "L 96.578246 -72.014734 \n", "L 96.997269 -72.318653 \n", "L 97.416293 -72.624853 \n", "L 97.835316 -72.933341 \n", "L 98.254339 -73.244121 \n", "L 98.673363 -73.557201 \n", "L 99.092386 -73.872595 \n", "L 99.511409 -74.190314 \n", "L 99.930433 -74.510372 \n", "L 100.349456 -74.832766 \n", "L 100.76848 -75.157506 \n", "L 101.187503 -75.48462 \n", "L 101.606526 -75.814071 \n", "L 102.02555 -76.145923 \n", "L 102.444573 -76.480137 \n", "L 102.863596 -76.816757 \n", "L 103.28262 -77.155756 \n", "L 103.701643 -77.497171 \n", "L 104.120677 -77.841005 \n", "L 104.5397 -78.187259 \n", "L 104.958723 -78.535929 \n", "L 105.377747 -78.887013 \n", "L 105.79677 -79.240554 \n", "L 106.215794 -79.596555 \n", "L 106.634817 -79.954966 \n", "L 107.05384 -80.31587 \n", "L 107.472864 -80.679224 \n", "L 107.891887 -81.045044 \n", "L 108.31091 -81.413338 \n", "L 108.729934 -81.7841 \n", "L 109.148957 -82.15736 \n", "L 109.567981 -82.533102 \n", "L 109.987004 -82.911318 \n", "L 110.406027 -83.292039 \n", "L 110.825051 -83.675275 \n", "L 111.244074 -84.060989 \n", "L 111.663097 -84.449235 \n", "L 112.082121 -84.839956 \n", "L 112.501144 -85.233209 \n", "L 112.920168 -85.628971 \n", "L 113.339191 -86.027235 \n", "L 113.758214 -86.428031 \n", "L 114.177248 -86.83135 \n", "L 114.596271 -87.23718 \n", "L 115.015294 -87.645519 \n", "L 115.434318 -88.056395 \n", "L 115.853341 -88.46979 \n", "L 116.272365 -88.885691 \n", "L 116.691388 -89.304124 \n", "L 117.110411 -89.725049 \n", "L 117.529435 -90.14852 \n", "L 117.948458 -90.574491 \n", "L 118.367482 -91.002995 \n", "L 118.786505 -91.433987 \n", "L 119.205528 -91.867497 \n", "L 119.624552 -92.3035 \n", "L 120.043575 -92.742007 \n", "L 120.462598 -93.183016 \n", "L 120.881622 -93.626534 \n", "L 121.300645 -94.072509 \n", "L 121.719669 -94.521002 \n", "L 122.138692 -94.971956 \n", "L 122.557715 -95.425379 \n", "L 122.976739 -95.881266 \n", "L 123.395762 -96.339601 \n", "L 123.814785 -96.8004 \n", "L 124.233819 -97.263638 \n", "L 124.652842 -97.729304 \n", "L 125.071866 -98.197412 \n", "L 125.490889 -98.667913 \n", "L 125.909912 -99.140829 \n", "L 126.328936 -99.616166 \n", "L 126.747959 -100.093858 \n", "L 127.166982 -100.57393 \n", "L 127.586006 -101.056359 \n", "L 128.005029 -101.541172 \n", "L 128.424053 -102.028269 \n", "L 128.843076 -102.517709 \n", "L 129.262099 -103.009497 \n", "L 129.681123 -103.503525 \n", "L 129.681123 -37.55625 \n", "L 129.681123 -37.55625 \n", "L 129.262099 -37.55625 \n", "L 128.843076 -37.55625 \n", "L 128.424053 -37.55625 \n", "L 128.005029 -37.55625 \n", "L 127.586006 -37.55625 \n", "L 127.166982 -37.55625 \n", "L 126.747959 -37.55625 \n", "L 126.328936 -37.55625 \n", "L 125.909912 -37.55625 \n", "L 125.490889 -37.55625 \n", "L 125.071866 -37.55625 \n", "L 124.652842 -37.55625 \n", "L 124.233819 -37.55625 \n", "L 123.814785 -37.55625 \n", "L 123.395762 -37.55625 \n", "L 122.976739 -37.55625 \n", "L 122.557715 -37.55625 \n", "L 122.138692 -37.55625 \n", "L 121.719669 -37.55625 \n", "L 121.300645 -37.55625 \n", "L 120.881622 -37.55625 \n", "L 120.462598 -37.55625 \n", "L 120.043575 -37.55625 \n", "L 119.624552 -37.55625 \n", "L 119.205528 -37.55625 \n", "L 118.786505 -37.55625 \n", "L 118.367482 -37.55625 \n", "L 117.948458 -37.55625 \n", "L 117.529435 -37.55625 \n", "L 117.110411 -37.55625 \n", "L 116.691388 -37.55625 \n", "L 116.272365 -37.55625 \n", "L 115.853341 -37.55625 \n", "L 115.434318 -37.55625 \n", "L 115.015294 -37.55625 \n", "L 114.596271 -37.55625 \n", "L 114.177248 -37.55625 \n", "L 113.758214 -37.55625 \n", "L 113.339191 -37.55625 \n", "L 112.920168 -37.55625 \n", "L 112.501144 -37.55625 \n", "L 112.082121 -37.55625 \n", "L 111.663097 -37.55625 \n", "L 111.244074 -37.55625 \n", "L 110.825051 -37.55625 \n", "L 110.406027 -37.55625 \n", "L 109.987004 -37.55625 \n", "L 109.567981 -37.55625 \n", "L 109.148957 -37.55625 \n", "L 108.729934 -37.55625 \n", "L 108.31091 -37.55625 \n", "L 107.891887 -37.55625 \n", "L 107.472864 -37.55625 \n", "L 107.05384 -37.55625 \n", "L 106.634817 -37.55625 \n", "L 106.215794 -37.55625 \n", "L 105.79677 -37.55625 \n", "L 105.377747 -37.55625 \n", "L 104.958723 -37.55625 \n", "L 104.5397 -37.55625 \n", "L 104.120677 -37.55625 \n", "L 103.701643 -37.55625 \n", "L 103.28262 -37.55625 \n", "L 102.863596 -37.55625 \n", "L 102.444573 -37.55625 \n", "L 102.02555 -37.55625 \n", "L 101.606526 -37.55625 \n", "L 101.187503 -37.55625 \n", "L 100.76848 -37.55625 \n", "L 100.349456 -37.55625 \n", "L 99.930433 -37.55625 \n", "L 99.511409 -37.55625 \n", "L 99.092386 -37.55625 \n", "L 98.673363 -37.55625 \n", "L 98.254339 -37.55625 \n", "L 97.835316 -37.55625 \n", "L 97.416293 -37.55625 \n", "L 96.997269 -37.55625 \n", "L 96.578246 -37.55625 \n", "L 96.159222 -37.55625 \n", "L 95.740199 -37.55625 \n", "L 95.321176 -37.55625 \n", "L 94.902152 -37.55625 \n", "L 94.483129 -37.55625 \n", "L 94.064106 -37.55625 \n", "L 93.645072 -37.55625 \n", "L 93.226049 -37.55625 \n", "L 92.807025 -37.55625 \n", "L 92.388002 -37.55625 \n", "L 91.968979 -37.55625 \n", "L 91.549955 -37.55625 \n", "L 91.130932 -37.55625 \n", "L 90.711908 -37.55625 \n", "L 90.292885 -37.55625 \n", "L 89.873862 -37.55625 \n", "L 89.454838 -37.55625 \n", "L 89.035815 -37.55625 \n", "L 88.616792 -37.55625 \n", "L 88.197768 -37.55625 \n", "L 87.778745 -37.55625 \n", "L 87.359721 -37.55625 \n", "L 86.940698 -37.55625 \n", "L 86.521675 -37.55625 \n", "L 86.102651 -37.55625 \n", "L 85.683628 -37.55625 \n", "L 85.264605 -37.55625 \n", "L 84.845581 -37.55625 \n", "L 84.426558 -37.55625 \n", "L 84.007534 -37.55625 \n", "L 83.588511 -37.55625 \n", "L 83.169488 -37.55625 \n", "L 82.750464 -37.55625 \n", "L 82.331441 -37.55625 \n", "L 81.912418 -37.55625 \n", "L 81.493394 -37.55625 \n", "L 81.074371 -37.55625 \n", "L 80.655347 -37.55625 \n", "L 80.236314 -37.55625 \n", "L 79.817291 -37.55625 \n", "L 79.398267 -37.55625 \n", "L 78.979244 -37.55625 \n", "L 78.56022 -37.55625 \n", "L 78.141197 -37.55625 \n", "L 77.722174 -37.55625 \n", "L 77.30315 -37.55625 \n", "L 76.884127 -37.55625 \n", "L 76.465104 -37.55625 \n", "L 76.04608 -37.55625 \n", "L 75.627057 -37.55625 \n", "L 75.208033 -37.55625 \n", "L 74.78901 -37.55625 \n", "L 74.369987 -37.55625 \n", "L 73.950963 -37.55625 \n", "L 73.53194 -37.55625 \n", "L 73.112917 -37.55625 \n", "L 72.693893 -37.55625 \n", "L 72.27487 -37.55625 \n", "L 71.855846 -37.55625 \n", "L 71.436823 -37.55625 \n", "L 71.0178 -37.55625 \n", "L 70.598776 -37.55625 \n", "L 70.179743 -37.55625 \n", "L 69.76072 -37.55625 \n", "L 69.341696 -37.55625 \n", "L 68.922673 -37.55625 \n", "L 68.503649 -37.55625 \n", "L 68.084626 -37.55625 \n", "L 67.665603 -37.55625 \n", "L 67.246579 -37.55625 \n", "L 66.827556 -37.55625 \n", "L 66.408532 -37.55625 \n", "L 65.989509 -37.55625 \n", "L 65.570486 -37.55625 \n", "L 65.151462 -37.55625 \n", "L 64.732439 -37.55625 \n", "L 64.313416 -37.55625 \n", "L 63.894392 -37.55625 \n", "L 63.475369 -37.55625 \n", "L 63.056345 -37.55625 \n", "L 62.637322 -37.55625 \n", "L 62.218299 -37.55625 \n", "L 61.799275 -37.55625 \n", "L 61.380252 -37.55625 \n", "L 60.961229 -37.55625 \n", "L 60.542205 -37.55625 \n", "L 60.123172 -37.55625 \n", "L 59.704148 -37.55625 \n", "L 59.285125 -37.55625 \n", "L 58.866102 -37.55625 \n", "L 58.447078 -37.55625 \n", "L 58.028055 -37.55625 \n", "L 57.609032 -37.55625 \n", "L 57.190008 -37.55625 \n", "L 56.770985 -37.55625 \n", "L 56.351961 -37.55625 \n", "L 55.932938 -37.55625 \n", "L 55.513915 -37.55625 \n", "L 55.094891 -37.55625 \n", "L 54.675868 -37.55625 \n", "L 54.256844 -37.55625 \n", "L 53.837821 -37.55625 \n", "L 53.418798 -37.55625 \n", "L 52.999774 -37.55625 \n", "L 52.580751 -37.55625 \n", "L 52.161728 -37.55625 \n", "L 51.742704 -37.55625 \n", "L 51.323681 -37.55625 \n", "L 50.904657 -37.55625 \n", "L 50.485634 -37.55625 \n", "L 50.066601 -37.55625 \n", "L 49.647577 -37.55625 \n", "L 49.228554 -37.55625 \n", "L 48.809531 -37.55625 \n", "L 48.390507 -37.55625 \n", "L 47.971484 -37.55625 \n", "L 47.55246 -37.55625 \n", "L 47.133437 -37.55625 \n", "L 46.714414 -37.55625 \n", "L 46.29539 -37.55625 \n", "L 45.876367 -37.55625 \n", "L 45.457344 -37.55625 \n", "L 45.03832 -37.55625 \n", "L 44.619297 -37.55625 \n", "L 44.200273 -37.55625 \n", "L 43.78125 -37.55625 \n", "z\n", "\" style=\"stroke: #1f77b4; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pc8b7a1e7e3)\">\n", "     <use xlink:href=\"#mbc49098ec0\" x=\"0\" y=\"226.194375\" style=\"fill: #1f77b4; fill-opacity: 0.5; stroke: #1f77b4; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_2\">\n", "    <defs>\n", "     <path id=\"m7d8159ae1c\" d=\"M 130.100146 -37.55625 \n", "L 130.100146 -103.999859 \n", "L 130.519174 -104.498474 \n", "L 130.938198 -104.999319 \n", "L 131.357221 -105.50239 \n", "L 131.776245 -106.007679 \n", "L 132.195268 -106.515148 \n", "L 132.614291 -107.024821 \n", "L 133.033315 -107.536625 \n", "L 133.452338 -108.050623 \n", "L 133.871367 -108.566699 \n", "L 134.290385 -109.084884 \n", "L 134.709408 -109.605133 \n", "L 135.128432 -110.127436 \n", "L 135.547455 -110.651785 \n", "L 135.966478 -111.178162 \n", "L 136.385502 -111.706494 \n", "L 136.804525 -112.236801 \n", "L 137.223554 -112.769035 \n", "L 137.642577 -113.303185 \n", "L 138.0616 -113.839214 \n", "L 138.480624 -114.377082 \n", "L 138.899647 -114.916806 \n", "L 139.31867 -115.458314 \n", "L 139.737694 -116.001589 \n", "L 140.156717 -116.546603 \n", "L 140.575746 -117.09332 \n", "L 140.994769 -117.641759 \n", "L 141.413792 -118.191784 \n", "L 141.832816 -118.743458 \n", "L 142.251839 -119.296692 \n", "L 142.670862 -119.851484 \n", "L 143.089886 -120.40779 \n", "L 143.508909 -120.965566 \n", "L 143.927938 -121.524792 \n", "L 144.346956 -122.085398 \n", "L 144.765979 -122.647373 \n", "L 145.185003 -123.210664 \n", "L 145.604026 -123.77527 \n", "L 146.02305 -124.341103 \n", "L 146.442073 -124.908143 \n", "L 146.861096 -125.476354 \n", "L 147.280125 -126.04572 \n", "L 147.699148 -126.61613 \n", "L 148.118171 -127.187586 \n", "L 148.537195 -127.760024 \n", "L 148.956218 -128.333409 \n", "L 149.375242 -128.907722 \n", "L 149.794265 -129.482882 \n", "L 150.213288 -130.058835 \n", "L 150.632317 -130.635572 \n", "L 151.051335 -131.213003 \n", "L 151.470358 -131.791083 \n", "L 151.889382 -132.369812 \n", "L 152.308405 -132.949073 \n", "L 152.727429 -133.528874 \n", "L 153.146452 -134.109117 \n", "L 153.565475 -134.689757 \n", "L 153.984504 -135.270784 \n", "L 154.403527 -135.85209 \n", "L 154.82255 -136.433649 \n", "L 155.241574 -137.015397 \n", "L 155.660597 -137.597253 \n", "L 156.079621 -138.179199 \n", "L 156.498644 -138.761181 \n", "L 156.917667 -139.3431 \n", "L 157.336696 -139.92492 \n", "L 157.755719 -140.506587 \n", "L 158.174743 -141.088029 \n", "L 158.593766 -141.669209 \n", "L 159.012789 -142.250011 \n", "L 159.431813 -142.830434 \n", "L 159.850836 -143.410388 \n", "L 160.269859 -143.989802 \n", "L 160.688888 -144.56863 \n", "L 161.107906 -145.146809 \n", "L 161.52693 -145.724258 \n", "L 161.945953 -146.300896 \n", "L 162.364976 -146.876687 \n", "L 162.784 -147.45155 \n", "L 163.203023 -148.025439 \n", "L 163.622046 -148.598265 \n", "L 164.041075 -149.169946 \n", "L 164.460098 -149.740447 \n", "L 164.879122 -150.309668 \n", "L 165.298145 -150.877564 \n", "L 165.298145 -37.55625 \n", "L 165.298145 -37.55625 \n", "L 164.879122 -37.55625 \n", "L 164.460098 -37.55625 \n", "L 164.041075 -37.55625 \n", "L 163.622046 -37.55625 \n", "L 163.203023 -37.55625 \n", "L 162.784 -37.55625 \n", "L 162.364976 -37.55625 \n", "L 161.945953 -37.55625 \n", "L 161.52693 -37.55625 \n", "L 161.107906 -37.55625 \n", "L 160.688888 -37.55625 \n", "L 160.269859 -37.55625 \n", "L 159.850836 -37.55625 \n", "L 159.431813 -37.55625 \n", "L 159.012789 -37.55625 \n", "L 158.593766 -37.55625 \n", "L 158.174743 -37.55625 \n", "L 157.755719 -37.55625 \n", "L 157.336696 -37.55625 \n", "L 156.917667 -37.55625 \n", "L 156.498644 -37.55625 \n", "L 156.079621 -37.55625 \n", "L 155.660597 -37.55625 \n", "L 155.241574 -37.55625 \n", "L 154.82255 -37.55625 \n", "L 154.403527 -37.55625 \n", "L 153.984504 -37.55625 \n", "L 153.565475 -37.55625 \n", "L 153.146452 -37.55625 \n", "L 152.727429 -37.55625 \n", "L 152.308405 -37.55625 \n", "L 151.889382 -37.55625 \n", "L 151.470358 -37.55625 \n", "L 151.051335 -37.55625 \n", "L 150.632317 -37.55625 \n", "L 150.213288 -37.55625 \n", "L 149.794265 -37.55625 \n", "L 149.375242 -37.55625 \n", "L 148.956218 -37.55625 \n", "L 148.537195 -37.55625 \n", "L 148.118171 -37.55625 \n", "L 147.699148 -37.55625 \n", "L 147.280125 -37.55625 \n", "L 146.861096 -37.55625 \n", "L 146.442073 -37.55625 \n", "L 146.02305 -37.55625 \n", "L 145.604026 -37.55625 \n", "L 145.185003 -37.55625 \n", "L 144.765979 -37.55625 \n", "L 144.346956 -37.55625 \n", "L 143.927938 -37.55625 \n", "L 143.508909 -37.55625 \n", "L 143.089886 -37.55625 \n", "L 142.670862 -37.55625 \n", "L 142.251839 -37.55625 \n", "L 141.832816 -37.55625 \n", "L 141.413792 -37.55625 \n", "L 140.994769 -37.55625 \n", "L 140.575746 -37.55625 \n", "L 140.156717 -37.55625 \n", "L 139.737694 -37.55625 \n", "L 139.31867 -37.55625 \n", "L 138.899647 -37.55625 \n", "L 138.480624 -37.55625 \n", "L 138.0616 -37.55625 \n", "L 137.642577 -37.55625 \n", "L 137.223554 -37.55625 \n", "L 136.804525 -37.55625 \n", "L 136.385502 -37.55625 \n", "L 135.966478 -37.55625 \n", "L 135.547455 -37.55625 \n", "L 135.128432 -37.55625 \n", "L 134.709408 -37.55625 \n", "L 134.290385 -37.55625 \n", "L 133.871367 -37.55625 \n", "L 133.452338 -37.55625 \n", "L 133.033315 -37.55625 \n", "L 132.614291 -37.55625 \n", "L 132.195268 -37.55625 \n", "L 131.776245 -37.55625 \n", "L 131.357221 -37.55625 \n", "L 130.938198 -37.55625 \n", "L 130.519174 -37.55625 \n", "L 130.100146 -37.55625 \n", "z\n", "\" style=\"stroke: #ff7f0e; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pc8b7a1e7e3)\">\n", "     <use xlink:href=\"#m7d8159ae1c\" x=\"0\" y=\"226.194375\" style=\"fill: #ff7f0e; fill-opacity: 0.5; stroke: #ff7f0e; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_3\">\n", "    <defs>\n", "     <path id=\"md985cd0a96\" d=\"M 165.717168 -37.55625 \n", "L 165.717168 -151.444028 \n", "L 166.136192 -152.00904 \n", "L 166.555215 -152.572493 \n", "L 166.974238 -153.134306 \n", "L 167.393267 -153.694452 \n", "L 167.81229 -154.252822 \n", "L 168.231314 -154.809345 \n", "L 168.650337 -155.363948 \n", "L 169.06936 -155.916595 \n", "L 169.488386 -156.467161 \n", "L 169.90741 -157.015591 \n", "L 170.326433 -157.561794 \n", "L 170.745456 -158.105727 \n", "L 171.16448 -158.647308 \n", "L 171.583503 -159.18641 \n", "L 172.002527 -159.723033 \n", "L 172.42155 -160.257071 \n", "L 172.840576 -160.788431 \n", "L 173.259599 -161.317034 \n", "L 173.678623 -161.842843 \n", "L 174.097646 -162.365732 \n", "L 174.516669 -162.885684 \n", "L 174.935693 -163.402553 \n", "L 175.354716 -163.916303 \n", "L 175.773739 -164.426864 \n", "L 176.192765 -164.934126 \n", "L 176.611789 -165.438008 \n", "L 177.030812 -165.938493 \n", "L 177.449835 -166.435463 \n", "L 177.868859 -166.928846 \n", "L 178.287882 -167.418534 \n", "L 178.706906 -167.904482 \n", "L 179.125929 -168.386609 \n", "L 179.544955 -168.864879 \n", "L 179.963978 -169.339147 \n", "L 180.383002 -169.809359 \n", "L 180.802025 -170.275463 \n", "L 181.221048 -170.73733 \n", "L 181.640072 -171.194944 \n", "L 182.059095 -171.648213 \n", "L 182.478119 -172.097022 \n", "L 182.897144 -172.541352 \n", "L 183.316168 -172.981103 \n", "L 183.735191 -173.416204 \n", "L 184.154215 -173.846556 \n", "L 184.57324 -174.272131 \n", "L 184.992264 -174.692803 \n", "L 185.411287 -175.108573 \n", "L 185.830311 -175.519314 \n", "L 186.249336 -175.924936 \n", "L 186.66836 -176.325448 \n", "L 187.087383 -176.720716 \n", "L 187.506407 -177.110692 \n", "L 187.92543 -177.49528 \n", "L 188.344453 -177.874478 \n", "L 188.763477 -178.248152 \n", "L 189.1825 -178.616238 \n", "L 189.601526 -178.978728 \n", "L 189.601526 -37.55625 \n", "L 189.601526 -37.55625 \n", "L 189.1825 -37.55625 \n", "L 188.763477 -37.55625 \n", "L 188.344453 -37.55625 \n", "L 187.92543 -37.55625 \n", "L 187.506407 -37.55625 \n", "L 187.087383 -37.55625 \n", "L 186.66836 -37.55625 \n", "L 186.249336 -37.55625 \n", "L 185.830311 -37.55625 \n", "L 185.411287 -37.55625 \n", "L 184.992264 -37.55625 \n", "L 184.57324 -37.55625 \n", "L 184.154215 -37.55625 \n", "L 183.735191 -37.55625 \n", "L 183.316168 -37.55625 \n", "L 182.897144 -37.55625 \n", "L 182.478119 -37.55625 \n", "L 182.059095 -37.55625 \n", "L 181.640072 -37.55625 \n", "L 181.221048 -37.55625 \n", "L 180.802025 -37.55625 \n", "L 180.383002 -37.55625 \n", "L 179.963978 -37.55625 \n", "L 179.544955 -37.55625 \n", "L 179.125929 -37.55625 \n", "L 178.706906 -37.55625 \n", "L 178.287882 -37.55625 \n", "L 177.868859 -37.55625 \n", "L 177.449835 -37.55625 \n", "L 177.030812 -37.55625 \n", "L 176.611789 -37.55625 \n", "L 176.192765 -37.55625 \n", "L 175.773739 -37.55625 \n", "L 175.354716 -37.55625 \n", "L 174.935693 -37.55625 \n", "L 174.516669 -37.55625 \n", "L 174.097646 -37.55625 \n", "L 173.678623 -37.55625 \n", "L 173.259599 -37.55625 \n", "L 172.840576 -37.55625 \n", "L 172.42155 -37.55625 \n", "L 172.002527 -37.55625 \n", "L 171.583503 -37.55625 \n", "L 171.16448 -37.55625 \n", "L 170.745456 -37.55625 \n", "L 170.326433 -37.55625 \n", "L 169.90741 -37.55625 \n", "L 169.488386 -37.55625 \n", "L 169.06936 -37.55625 \n", "L 168.650337 -37.55625 \n", "L 168.231314 -37.55625 \n", "L 167.81229 -37.55625 \n", "L 167.393267 -37.55625 \n", "L 166.974238 -37.55625 \n", "L 166.555215 -37.55625 \n", "L 166.136192 -37.55625 \n", "L 165.717168 -37.55625 \n", "z\n", "\" style=\"stroke: #2ca02c; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pc8b7a1e7e3)\">\n", "     <use xlink:href=\"#md985cd0a96\" x=\"0\" y=\"226.194375\" style=\"fill: #2ca02c; fill-opacity: 0.5; stroke: #2ca02c; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_4\">\n", "    <defs>\n", "     <path id=\"m0df0cce004\" d=\"M 190.020549 -37.55625 \n", "L 190.020549 -179.335495 \n", "L 190.439573 -179.686494 \n", "L 190.858596 -180.031671 \n", "L 191.277621 -180.370972 \n", "L 191.696644 -180.704307 \n", "L 192.115669 -181.031604 \n", "L 192.534692 -181.352863 \n", "L 192.953716 -181.668002 \n", "L 193.37274 -181.976914 \n", "L 193.791764 -182.279562 \n", "L 194.210787 -182.575938 \n", "L 194.62981 -182.865952 \n", "L 195.048834 -183.14953 \n", "L 195.467858 -183.426656 \n", "L 195.886882 -183.697257 \n", "L 196.305905 -183.96127 \n", "L 196.72493 -184.218668 \n", "L 197.143953 -184.469415 \n", "L 197.562976 -184.713421 \n", "L 197.982001 -184.95065 \n", "L 198.401024 -185.181074 \n", "L 198.820049 -185.404641 \n", "L 199.239072 -185.621303 \n", "L 199.658096 -185.831036 \n", "L 200.07712 -186.033775 \n", "L 200.496144 -186.229484 \n", "L 200.915167 -186.418146 \n", "L 201.334191 -186.599706 \n", "L 201.753215 -186.774138 \n", "L 202.172239 -186.941387 \n", "L 202.591263 -187.101444 \n", "L 203.010287 -187.254282 \n", "L 203.42931 -187.399829 \n", "L 203.848334 -187.538113 \n", "L 204.267357 -187.66907 \n", "L 204.686381 -187.792681 \n", "L 205.105405 -187.908921 \n", "L 205.524429 -188.017779 \n", "L 205.943453 -188.119221 \n", "L 206.362476 -188.213228 \n", "L 206.7815 -188.299808 \n", "L 207.200524 -188.378872 \n", "L 207.619548 -188.450483 \n", "L 208.038572 -188.514587 \n", "L 208.457595 -188.571184 \n", "L 208.876619 -188.620229 \n", "L 209.295643 -188.661775 \n", "L 209.714667 -188.695769 \n", "L 210.13369 -188.722238 \n", "L 210.552714 -188.741127 \n", "L 210.971738 -188.752465 \n", "L 210.971738 -37.55625 \n", "L 210.971738 -37.55625 \n", "L 210.552714 -37.55625 \n", "L 210.13369 -37.55625 \n", "L 209.714667 -37.55625 \n", "L 209.295643 -37.55625 \n", "L 208.876619 -37.55625 \n", "L 208.457595 -37.55625 \n", "L 208.038572 -37.55625 \n", "L 207.619548 -37.55625 \n", "L 207.200524 -37.55625 \n", "L 206.7815 -37.55625 \n", "L 206.362476 -37.55625 \n", "L 205.943453 -37.55625 \n", "L 205.524429 -37.55625 \n", "L 205.105405 -37.55625 \n", "L 204.686381 -37.55625 \n", "L 204.267357 -37.55625 \n", "L 203.848334 -37.55625 \n", "L 203.42931 -37.55625 \n", "L 203.010287 -37.55625 \n", "L 202.591263 -37.55625 \n", "L 202.172239 -37.55625 \n", "L 201.753215 -37.55625 \n", "L 201.334191 -37.55625 \n", "L 200.915167 -37.55625 \n", "L 200.496144 -37.55625 \n", "L 200.07712 -37.55625 \n", "L 199.658096 -37.55625 \n", "L 199.239072 -37.55625 \n", "L 198.820049 -37.55625 \n", "L 198.401024 -37.55625 \n", "L 197.982001 -37.55625 \n", "L 197.562976 -37.55625 \n", "L 197.143953 -37.55625 \n", "L 196.72493 -37.55625 \n", "L 196.305905 -37.55625 \n", "L 195.886882 -37.55625 \n", "L 195.467858 -37.55625 \n", "L 195.048834 -37.55625 \n", "L 194.62981 -37.55625 \n", "L 194.210787 -37.55625 \n", "L 193.791764 -37.55625 \n", "L 193.37274 -37.55625 \n", "L 192.953716 -37.55625 \n", "L 192.534692 -37.55625 \n", "L 192.115669 -37.55625 \n", "L 191.696644 -37.55625 \n", "L 191.277621 -37.55625 \n", "L 190.858596 -37.55625 \n", "L 190.439573 -37.55625 \n", "L 190.020549 -37.55625 \n", "z\n", "\" style=\"stroke: #d62728; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pc8b7a1e7e3)\">\n", "     <use xlink:href=\"#m0df0cce004\" x=\"0\" y=\"226.194375\" style=\"fill: #d62728; fill-opacity: 0.5; stroke: #d62728; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_5\">\n", "    <defs>\n", "     <path id=\"m227058dd92\" d=\"M 211.390762 -37.55625 \n", "L 211.390762 -188.75625 \n", "L 211.809785 -188.752465 \n", "L 212.228809 -188.741127 \n", "L 212.647833 -188.722238 \n", "L 213.066857 -188.695787 \n", "L 213.485881 -188.661793 \n", "L 213.904904 -188.620265 \n", "L 214.323928 -188.571184 \n", "L 214.742952 -188.514587 \n", "L 215.161976 -188.450483 \n", "L 215.581 -188.378872 \n", "L 216.000023 -188.29979 \n", "L 216.419047 -188.213228 \n", "L 216.838071 -188.119221 \n", "L 217.257095 -188.017779 \n", "L 217.676118 -187.908921 \n", "L 218.095142 -187.792699 \n", "L 218.514166 -187.66907 \n", "L 218.933189 -187.538113 \n", "L 219.352213 -187.399829 \n", "L 219.771237 -187.254264 \n", "L 220.190261 -187.101444 \n", "L 220.609285 -186.941387 \n", "L 221.028308 -186.77412 \n", "L 221.447332 -186.599706 \n", "L 221.866356 -186.418146 \n", "L 222.28538 -186.229484 \n", "L 222.704403 -186.033775 \n", "L 223.123428 -185.831036 \n", "L 223.542451 -185.621303 \n", "L 223.961474 -185.404641 \n", "L 224.380499 -185.181074 \n", "L 224.799522 -184.95065 \n", "L 225.218546 -184.713421 \n", "L 225.63757 -184.469397 \n", "L 226.056594 -184.218686 \n", "L 226.475617 -183.96127 \n", "L 226.894642 -183.697257 \n", "L 227.313665 -183.426656 \n", "L 227.732688 -183.14953 \n", "L 228.151713 -182.865952 \n", "L 228.570736 -182.575947 \n", "L 228.989761 -182.27958 \n", "L 229.408784 -181.976914 \n", "L 229.827808 -181.667984 \n", "L 230.246833 -181.352863 \n", "L 230.665856 -181.031622 \n", "L 231.084879 -180.704298 \n", "L 231.503903 -180.370972 \n", "L 231.922926 -180.031671 \n", "L 232.341951 -179.686494 \n", "L 232.760974 -179.335495 \n", "L 232.760974 -37.55625 \n", "L 232.760974 -37.55625 \n", "L 232.341951 -37.55625 \n", "L 231.922926 -37.55625 \n", "L 231.503903 -37.55625 \n", "L 231.084879 -37.55625 \n", "L 230.665856 -37.55625 \n", "L 230.246833 -37.55625 \n", "L 229.827808 -37.55625 \n", "L 229.408784 -37.55625 \n", "L 228.989761 -37.55625 \n", "L 228.570736 -37.55625 \n", "L 228.151713 -37.55625 \n", "L 227.732688 -37.55625 \n", "L 227.313665 -37.55625 \n", "L 226.894642 -37.55625 \n", "L 226.475617 -37.55625 \n", "L 226.056594 -37.55625 \n", "L 225.63757 -37.55625 \n", "L 225.218546 -37.55625 \n", "L 224.799522 -37.55625 \n", "L 224.380499 -37.55625 \n", "L 223.961474 -37.55625 \n", "L 223.542451 -37.55625 \n", "L 223.123428 -37.55625 \n", "L 222.704403 -37.55625 \n", "L 222.28538 -37.55625 \n", "L 221.866356 -37.55625 \n", "L 221.447332 -37.55625 \n", "L 221.028308 -37.55625 \n", "L 220.609285 -37.55625 \n", "L 220.190261 -37.55625 \n", "L 219.771237 -37.55625 \n", "L 219.352213 -37.55625 \n", "L 218.933189 -37.55625 \n", "L 218.514166 -37.55625 \n", "L 218.095142 -37.55625 \n", "L 217.676118 -37.55625 \n", "L 217.257095 -37.55625 \n", "L 216.838071 -37.55625 \n", "L 216.419047 -37.55625 \n", "L 216.000023 -37.55625 \n", "L 215.581 -37.55625 \n", "L 215.161976 -37.55625 \n", "L 214.742952 -37.55625 \n", "L 214.323928 -37.55625 \n", "L 213.904904 -37.55625 \n", "L 213.485881 -37.55625 \n", "L 213.066857 -37.55625 \n", "L 212.647833 -37.55625 \n", "L 212.228809 -37.55625 \n", "L 211.809785 -37.55625 \n", "L 211.390762 -37.55625 \n", "z\n", "\" style=\"stroke: #9467bd; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pc8b7a1e7e3)\">\n", "     <use xlink:href=\"#m227058dd92\" x=\"0\" y=\"226.194375\" style=\"fill: #9467bd; fill-opacity: 0.5; stroke: #9467bd; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_6\">\n", "    <defs>\n", "     <path id=\"m2920a56ada\" d=\"M 233.179997 -37.55625 \n", "L 233.179997 -178.97871 \n", "L 233.599021 -178.616238 \n", "L 234.018044 -178.248143 \n", "L 234.43707 -177.874478 \n", "L 234.856093 -177.495298 \n", "L 235.275117 -177.110692 \n", "L 235.69414 -176.720716 \n", "L 236.113164 -176.325448 \n", "L 236.532189 -175.924972 \n", "L 236.951213 -175.519296 \n", "L 237.370236 -175.108573 \n", "L 237.78926 -174.692812 \n", "L 238.208283 -174.272113 \n", "L 238.627306 -173.846556 \n", "L 239.04633 -173.416186 \n", "L 239.465353 -172.981085 \n", "L 239.884379 -172.541352 \n", "L 240.303402 -172.097022 \n", "L 240.722426 -171.648195 \n", "L 241.141449 -171.194944 \n", "L 241.560475 -170.73733 \n", "L 241.979498 -170.275445 \n", "L 242.398522 -169.809359 \n", "L 242.817545 -169.339147 \n", "L 243.236571 -168.864879 \n", "L 243.655594 -168.386636 \n", "L 244.074618 -167.904482 \n", "L 244.493641 -167.418534 \n", "L 244.912665 -166.928846 \n", "L 245.331688 -166.435463 \n", "L 245.750711 -165.938511 \n", "L 246.169735 -165.438035 \n", "L 246.588761 -164.934126 \n", "L 247.007784 -164.426855 \n", "L 247.426807 -163.916294 \n", "L 247.845831 -163.402553 \n", "L 248.264854 -162.885675 \n", "L 248.683877 -162.36575 \n", "L 249.102901 -161.842825 \n", "L 249.521924 -161.317034 \n", "L 249.94095 -160.788422 \n", "L 250.359973 -160.257053 \n", "L 250.778997 -159.723033 \n", "L 251.19802 -159.18641 \n", "L 251.617044 -158.647271 \n", "L 252.036067 -158.105727 \n", "L 252.45509 -157.561794 \n", "L 252.874114 -157.015573 \n", "L 253.29314 -156.467143 \n", "L 253.712163 -155.916577 \n", "L 254.131186 -155.363957 \n", "L 254.55021 -154.809345 \n", "L 254.969233 -154.252822 \n", "L 255.388257 -153.694452 \n", "L 255.80728 -153.134324 \n", "L 256.226303 -152.572493 \n", "L 256.645327 -152.00904 \n", "L 257.06435 -151.444037 \n", "L 257.06435 -37.55625 \n", "L 257.06435 -37.55625 \n", "L 256.645327 -37.55625 \n", "L 256.226303 -37.55625 \n", "L 255.80728 -37.55625 \n", "L 255.388257 -37.55625 \n", "L 254.969233 -37.55625 \n", "L 254.55021 -37.55625 \n", "L 254.131186 -37.55625 \n", "L 253.712163 -37.55625 \n", "L 253.29314 -37.55625 \n", "L 252.874114 -37.55625 \n", "L 252.45509 -37.55625 \n", "L 252.036067 -37.55625 \n", "L 251.617044 -37.55625 \n", "L 251.19802 -37.55625 \n", "L 250.778997 -37.55625 \n", "L 250.359973 -37.55625 \n", "L 249.94095 -37.55625 \n", "L 249.521924 -37.55625 \n", "L 249.102901 -37.55625 \n", "L 248.683877 -37.55625 \n", "L 248.264854 -37.55625 \n", "L 247.845831 -37.55625 \n", "L 247.426807 -37.55625 \n", "L 247.007784 -37.55625 \n", "L 246.588761 -37.55625 \n", "L 246.169735 -37.55625 \n", "L 245.750711 -37.55625 \n", "L 245.331688 -37.55625 \n", "L 244.912665 -37.55625 \n", "L 244.493641 -37.55625 \n", "L 244.074618 -37.55625 \n", "L 243.655594 -37.55625 \n", "L 243.236571 -37.55625 \n", "L 242.817545 -37.55625 \n", "L 242.398522 -37.55625 \n", "L 241.979498 -37.55625 \n", "L 241.560475 -37.55625 \n", "L 241.141449 -37.55625 \n", "L 240.722426 -37.55625 \n", "L 240.303402 -37.55625 \n", "L 239.884379 -37.55625 \n", "L 239.465353 -37.55625 \n", "L 239.04633 -37.55625 \n", "L 238.627306 -37.55625 \n", "L 238.208283 -37.55625 \n", "L 237.78926 -37.55625 \n", "L 237.370236 -37.55625 \n", "L 236.951213 -37.55625 \n", "L 236.532189 -37.55625 \n", "L 236.113164 -37.55625 \n", "L 235.69414 -37.55625 \n", "L 235.275117 -37.55625 \n", "L 234.856093 -37.55625 \n", "L 234.43707 -37.55625 \n", "L 234.018044 -37.55625 \n", "L 233.599021 -37.55625 \n", "L 233.179997 -37.55625 \n", "z\n", "\" style=\"stroke: #8c564b; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pc8b7a1e7e3)\">\n", "     <use xlink:href=\"#m2920a56ada\" x=\"0\" y=\"226.194375\" style=\"fill: #8c564b; fill-opacity: 0.5; stroke: #8c564b; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_7\">\n", "    <defs>\n", "     <path id=\"mc676ddd24f\" d=\"M 257.483373 -37.55625 \n", "L 257.483373 -150.877564 \n", "L 257.902402 -150.309659 \n", "L 258.321425 -149.740438 \n", "L 258.740449 -149.169937 \n", "L 259.159472 -148.598265 \n", "L 259.578495 -148.025439 \n", "L 259.997519 -147.451568 \n", "L 260.416542 -146.876687 \n", "L 260.835565 -146.300905 \n", "L 261.254594 -145.724249 \n", "L 261.673617 -145.146809 \n", "L 262.092641 -144.568648 \n", "L 262.511664 -143.98982 \n", "L 262.930687 -143.410397 \n", "L 263.349711 -142.830443 \n", "L 263.768734 -142.250038 \n", "L 264.187757 -141.6692 \n", "L 264.606786 -141.088029 \n", "L 265.025804 -140.506587 \n", "L 265.444828 -139.924929 \n", "L 265.863851 -139.3431 \n", "L 266.282874 -138.761181 \n", "L 266.701898 -138.179208 \n", "L 267.120921 -137.597262 \n", "L 267.539945 -137.015397 \n", "L 267.958973 -136.433649 \n", "L 268.377996 -135.852099 \n", "L 268.79702 -135.270784 \n", "L 269.216043 -134.689784 \n", "L 269.635066 -134.109126 \n", "L 270.05409 -133.528874 \n", "L 270.473113 -132.949091 \n", "L 270.892137 -132.369812 \n", "L 271.311165 -131.791092 \n", "L 271.730188 -131.212994 \n", "L 272.149212 -130.635563 \n", "L 272.568235 -130.058835 \n", "L 272.987258 -129.482864 \n", "L 273.406282 -128.907722 \n", "L 273.825305 -128.333409 \n", "L 274.244329 -127.760024 \n", "L 274.663357 -127.187568 \n", "L 275.082375 -126.61613 \n", "L 275.501399 -126.045711 \n", "L 275.920422 -125.476363 \n", "L 276.339445 -124.90817 \n", "L 276.758469 -124.341112 \n", "L 277.177492 -123.77527 \n", "L 277.596516 -123.210682 \n", "L 278.015544 -122.647364 \n", "L 278.434567 -122.08538 \n", "L 278.853591 -121.524774 \n", "L 279.272614 -120.965566 \n", "L 279.691638 -120.407781 \n", "L 280.110661 -119.851493 \n", "L 280.529684 -119.296692 \n", "L 280.948708 -118.743458 \n", "L 281.367736 -118.191784 \n", "L 281.786754 -117.641741 \n", "L 282.205778 -117.093338 \n", "L 282.624801 -116.546603 \n", "L 283.043825 -116.001589 \n", "L 283.462848 -115.458314 \n", "L 283.881871 -114.916806 \n", "L 284.300895 -114.3771 \n", "L 284.719923 -113.839196 \n", "L 285.138946 -113.303185 \n", "L 285.55797 -112.769035 \n", "L 285.976993 -112.236801 \n", "L 286.396017 -111.706494 \n", "L 286.81504 -111.178162 \n", "L 287.234063 -110.651798 \n", "L 287.653087 -110.127454 \n", "L 288.072115 -109.605133 \n", "L 288.491138 -109.084884 \n", "L 288.910162 -108.566699 \n", "L 289.329185 -108.050623 \n", "L 289.748209 -107.536643 \n", "L 290.167232 -107.024821 \n", "L 290.586255 -106.515166 \n", "L 291.005279 -106.007679 \n", "L 291.424307 -105.502372 \n", "L 291.843326 -104.999306 \n", "L 292.262349 -104.498456 \n", "L 292.681372 -103.999873 \n", "L 292.681372 -37.55625 \n", "L 292.681372 -37.55625 \n", "L 292.262349 -37.55625 \n", "L 291.843326 -37.55625 \n", "L 291.424307 -37.55625 \n", "L 291.005279 -37.55625 \n", "L 290.586255 -37.55625 \n", "L 290.167232 -37.55625 \n", "L 289.748209 -37.55625 \n", "L 289.329185 -37.55625 \n", "L 288.910162 -37.55625 \n", "L 288.491138 -37.55625 \n", "L 288.072115 -37.55625 \n", "L 287.653087 -37.55625 \n", "L 287.234063 -37.55625 \n", "L 286.81504 -37.55625 \n", "L 286.396017 -37.55625 \n", "L 285.976993 -37.55625 \n", "L 285.55797 -37.55625 \n", "L 285.138946 -37.55625 \n", "L 284.719923 -37.55625 \n", "L 284.300895 -37.55625 \n", "L 283.881871 -37.55625 \n", "L 283.462848 -37.55625 \n", "L 283.043825 -37.55625 \n", "L 282.624801 -37.55625 \n", "L 282.205778 -37.55625 \n", "L 281.786754 -37.55625 \n", "L 281.367736 -37.55625 \n", "L 280.948708 -37.55625 \n", "L 280.529684 -37.55625 \n", "L 280.110661 -37.55625 \n", "L 279.691638 -37.55625 \n", "L 279.272614 -37.55625 \n", "L 278.853591 -37.55625 \n", "L 278.434567 -37.55625 \n", "L 278.015544 -37.55625 \n", "L 277.596516 -37.55625 \n", "L 277.177492 -37.55625 \n", "L 276.758469 -37.55625 \n", "L 276.339445 -37.55625 \n", "L 275.920422 -37.55625 \n", "L 275.501399 -37.55625 \n", "L 275.082375 -37.55625 \n", "L 274.663357 -37.55625 \n", "L 274.244329 -37.55625 \n", "L 273.825305 -37.55625 \n", "L 273.406282 -37.55625 \n", "L 272.987258 -37.55625 \n", "L 272.568235 -37.55625 \n", "L 272.149212 -37.55625 \n", "L 271.730188 -37.55625 \n", "L 271.311165 -37.55625 \n", "L 270.892137 -37.55625 \n", "L 270.473113 -37.55625 \n", "L 270.05409 -37.55625 \n", "L 269.635066 -37.55625 \n", "L 269.216043 -37.55625 \n", "L 268.79702 -37.55625 \n", "L 268.377996 -37.55625 \n", "L 267.958973 -37.55625 \n", "L 267.539945 -37.55625 \n", "L 267.120921 -37.55625 \n", "L 266.701898 -37.55625 \n", "L 266.282874 -37.55625 \n", "L 265.863851 -37.55625 \n", "L 265.444828 -37.55625 \n", "L 265.025804 -37.55625 \n", "L 264.606786 -37.55625 \n", "L 264.187757 -37.55625 \n", "L 263.768734 -37.55625 \n", "L 263.349711 -37.55625 \n", "L 262.930687 -37.55625 \n", "L 262.511664 -37.55625 \n", "L 262.092641 -37.55625 \n", "L 261.673617 -37.55625 \n", "L 261.254594 -37.55625 \n", "L 260.835565 -37.55625 \n", "L 260.416542 -37.55625 \n", "L 259.997519 -37.55625 \n", "L 259.578495 -37.55625 \n", "L 259.159472 -37.55625 \n", "L 258.740449 -37.55625 \n", "L 258.321425 -37.55625 \n", "L 257.902402 -37.55625 \n", "L 257.483373 -37.55625 \n", "z\n", "\" style=\"stroke: #e377c2; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pc8b7a1e7e3)\">\n", "     <use xlink:href=\"#mc676ddd24f\" x=\"0\" y=\"226.194375\" style=\"fill: #e377c2; fill-opacity: 0.5; stroke: #e377c2; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_8\">\n", "    <defs>\n", "     <path id=\"m32cb1aa068\" d=\"M 293.100396 -37.55625 \n", "L 293.100396 -103.503543 \n", "L 293.519419 -103.009497 \n", "L 293.938442 -102.517727 \n", "L 294.357466 -102.028282 \n", "L 294.776494 -101.541154 \n", "L 295.195518 -101.056377 \n", "L 295.614541 -100.57393 \n", "L 296.033564 -100.093858 \n", "L 296.452588 -99.616166 \n", "L 296.871611 -99.140847 \n", "L 297.290634 -98.667927 \n", "L 297.709658 -98.197412 \n", "L 298.128681 -97.729317 \n", "L 298.547705 -97.263651 \n", "L 298.966728 -96.8004 \n", "L 299.385751 -96.339614 \n", "L 299.804775 -95.881266 \n", "L 300.223798 -95.425379 \n", "L 300.642821 -94.971956 \n", "L 301.061845 -94.521002 \n", "L 301.480868 -94.072522 \n", "L 301.899902 -93.626521 \n", "L 302.318925 -93.18303 \n", "L 302.737948 -92.742021 \n", "L 303.156972 -92.3035 \n", "L 303.575995 -91.867484 \n", "L 303.995018 -91.433973 \n", "L 304.414042 -91.002982 \n", "L 304.833065 -90.574491 \n", "L 305.252089 -90.14852 \n", "L 305.671112 -89.725049 \n", "L 306.090135 -89.304115 \n", "L 306.509159 -88.885691 \n", "L 306.928182 -88.469781 \n", "L 307.347206 -88.056395 \n", "L 307.766229 -87.645533 \n", "L 308.185252 -87.23718 \n", "L 308.604276 -86.83135 \n", "L 309.023299 -86.42804 \n", "L 309.442322 -86.027257 \n", "L 309.861346 -85.628985 \n", "L 310.280369 -85.233218 \n", "L 310.699393 -84.839969 \n", "L 311.118416 -84.449235 \n", "L 311.537439 -84.060998 \n", "L 311.956473 -83.675275 \n", "L 312.375496 -83.292048 \n", "L 312.794519 -82.911318 \n", "L 313.213543 -82.533093 \n", "L 313.632566 -82.15736 \n", "L 314.05159 -81.7841 \n", "L 314.470613 -81.413329 \n", "L 314.889636 -81.045044 \n", "L 315.30866 -80.679224 \n", "L 315.727683 -80.31587 \n", "L 316.146706 -79.95498 \n", "L 316.56573 -79.596541 \n", "L 316.984753 -79.240563 \n", "L 317.403777 -78.887022 \n", "L 317.8228 -78.535929 \n", "L 318.241823 -78.18725 \n", "L 318.660847 -77.841014 \n", "L 319.07987 -77.497185 \n", "L 319.498894 -77.155779 \n", "L 319.917917 -76.816766 \n", "L 320.33694 -76.480155 \n", "L 320.755964 -76.145932 \n", "L 321.174987 -75.814089 \n", "L 321.59401 -75.48462 \n", "L 322.013044 -75.157515 \n", "L 322.432067 -74.832766 \n", "L 322.851091 -74.510372 \n", "L 323.270114 -74.190323 \n", "L 323.689137 -73.872595 \n", "L 324.108161 -73.557201 \n", "L 324.527184 -73.244112 \n", "L 324.946207 -72.933334 \n", "L 325.365231 -72.624853 \n", "L 325.784254 -72.318653 \n", "L 326.203278 -72.014734 \n", "L 326.622301 -71.713082 \n", "L 327.041324 -71.413691 \n", "L 327.460348 -71.116547 \n", "L 327.879371 -70.821638 \n", "L 328.298394 -70.528959 \n", "L 328.717418 -70.238506 \n", "L 329.136441 -69.950255 \n", "L 329.555465 -69.664202 \n", "L 329.974488 -69.38034 \n", "L 330.393511 -69.098652 \n", "L 330.812535 -68.819135 \n", "L 331.231558 -68.541769 \n", "L 331.650582 -68.266558 \n", "L 332.069615 -67.993472 \n", "L 332.488638 -67.72251 \n", "L 332.907662 -67.453671 \n", "L 333.326685 -67.18693 \n", "L 333.745708 -66.922281 \n", "L 334.164732 -66.659719 \n", "L 334.583755 -66.399223 \n", "L 335.002779 -66.140786 \n", "L 335.421802 -65.884393 \n", "L 335.840825 -65.630046 \n", "L 336.259849 -65.377717 \n", "L 336.678872 -65.1274 \n", "L 337.097895 -64.879095 \n", "L 337.516919 -64.632773 \n", "L 337.935942 -64.388436 \n", "L 338.354966 -64.146071 \n", "L 338.773989 -63.905661 \n", "L 339.193012 -63.667193 \n", "L 339.612036 -63.430662 \n", "L 340.031059 -63.196054 \n", "L 340.450082 -62.963361 \n", "L 340.869106 -62.732564 \n", "L 341.288129 -62.503661 \n", "L 341.707153 -62.276627 \n", "L 342.126176 -62.051468 \n", "L 342.545199 -61.828161 \n", "L 342.964223 -61.606694 \n", "L 343.383246 -61.387058 \n", "L 343.80227 -61.169246 \n", "L 344.221293 -60.953234 \n", "L 344.640316 -60.739027 \n", "L 345.05934 -60.5266 \n", "L 345.478373 -60.315941 \n", "L 345.897396 -60.107054 \n", "L 346.31642 -59.899913 \n", "L 346.735443 -59.694515 \n", "L 347.154467 -59.490845 \n", "L 347.57349 -59.288886 \n", "L 347.992513 -59.088635 \n", "L 348.411537 -58.890075 \n", "L 348.83056 -58.693198 \n", "L 349.249583 -58.497991 \n", "L 349.668607 -58.304447 \n", "L 350.08763 -58.112546 \n", "L 350.506654 -57.922284 \n", "L 350.925677 -57.733647 \n", "L 351.3447 -57.546623 \n", "L 351.763724 -57.361203 \n", "L 352.182747 -57.177374 \n", "L 352.60177 -56.995122 \n", "L 353.020794 -56.814445 \n", "L 353.439817 -56.635318 \n", "L 353.858841 -56.457746 \n", "L 354.277864 -56.281706 \n", "L 354.696887 -56.10719 \n", "L 355.115911 -55.934187 \n", "L 355.534944 -55.762684 \n", "L 355.953968 -55.592677 \n", "L 356.372991 -55.424149 \n", "L 356.792014 -55.257094 \n", "L 357.211038 -55.0915 \n", "L 357.630061 -54.927351 \n", "L 358.049084 -54.764638 \n", "L 358.468108 -54.603354 \n", "L 358.887131 -54.44349 \n", "L 359.306155 -54.285028 \n", "L 359.725178 -54.127959 \n", "L 360.144201 -53.972279 \n", "L 360.563225 -53.817971 \n", "L 360.982248 -53.665026 \n", "L 361.401271 -53.513441 \n", "L 361.820295 -53.36319 \n", "L 362.239318 -53.21428 \n", "L 362.658342 -53.066687 \n", "L 363.077365 -52.920408 \n", "L 363.496388 -52.77543 \n", "L 363.915412 -52.631744 \n", "L 364.334435 -52.489343 \n", "L 364.753458 -52.348214 \n", "L 365.172482 -52.208345 \n", "L 365.591515 -52.069725 \n", "L 366.010539 -51.93235 \n", "L 366.429562 -51.796209 \n", "L 366.848585 -51.66129 \n", "L 367.267609 -51.527587 \n", "L 367.686632 -51.395085 \n", "L 368.105656 -51.263777 \n", "L 368.524679 -51.133654 \n", "L 368.943702 -51.004705 \n", "L 369.362726 -50.876921 \n", "L 369.781749 -50.750293 \n", "L 370.200772 -50.624814 \n", "L 370.619796 -50.500472 \n", "L 371.038819 -50.377256 \n", "L 371.457843 -50.255161 \n", "L 371.876866 -50.134174 \n", "L 372.295889 -50.01429 \n", "L 372.714913 -49.895495 \n", "L 373.133936 -49.777787 \n", "L 373.552959 -49.661149 \n", "L 373.971983 -49.54558 \n", "L 374.391006 -49.431065 \n", "L 374.81003 -49.317595 \n", "L 375.229053 -49.205167 \n", "L 375.648086 -49.093766 \n", "L 376.06711 -48.983392 \n", "L 376.486133 -48.874029 \n", "L 376.905156 -48.765669 \n", "L 377.32418 -48.658306 \n", "L 377.743203 -48.551931 \n", "L 378.162227 -48.446538 \n", "L 378.58125 -48.342115 \n", "L 378.58125 -37.55625 \n", "L 378.58125 -37.55625 \n", "L 378.162227 -37.55625 \n", "L 377.743203 -37.55625 \n", "L 377.32418 -37.55625 \n", "L 376.905156 -37.55625 \n", "L 376.486133 -37.55625 \n", "L 376.06711 -37.55625 \n", "L 375.648086 -37.55625 \n", "L 375.229053 -37.55625 \n", "L 374.81003 -37.55625 \n", "L 374.391006 -37.55625 \n", "L 373.971983 -37.55625 \n", "L 373.552959 -37.55625 \n", "L 373.133936 -37.55625 \n", "L 372.714913 -37.55625 \n", "L 372.295889 -37.55625 \n", "L 371.876866 -37.55625 \n", "L 371.457843 -37.55625 \n", "L 371.038819 -37.55625 \n", "L 370.619796 -37.55625 \n", "L 370.200772 -37.55625 \n", "L 369.781749 -37.55625 \n", "L 369.362726 -37.55625 \n", "L 368.943702 -37.55625 \n", "L 368.524679 -37.55625 \n", "L 368.105656 -37.55625 \n", "L 367.686632 -37.55625 \n", "L 367.267609 -37.55625 \n", "L 366.848585 -37.55625 \n", "L 366.429562 -37.55625 \n", "L 366.010539 -37.55625 \n", "L 365.591515 -37.55625 \n", "L 365.172482 -37.55625 \n", "L 364.753458 -37.55625 \n", "L 364.334435 -37.55625 \n", "L 363.915412 -37.55625 \n", "L 363.496388 -37.55625 \n", "L 363.077365 -37.55625 \n", "L 362.658342 -37.55625 \n", "L 362.239318 -37.55625 \n", "L 361.820295 -37.55625 \n", "L 361.401271 -37.55625 \n", "L 360.982248 -37.55625 \n", "L 360.563225 -37.55625 \n", "L 360.144201 -37.55625 \n", "L 359.725178 -37.55625 \n", "L 359.306155 -37.55625 \n", "L 358.887131 -37.55625 \n", "L 358.468108 -37.55625 \n", "L 358.049084 -37.55625 \n", "L 357.630061 -37.55625 \n", "L 357.211038 -37.55625 \n", "L 356.792014 -37.55625 \n", "L 356.372991 -37.55625 \n", "L 355.953968 -37.55625 \n", "L 355.534944 -37.55625 \n", "L 355.115911 -37.55625 \n", "L 354.696887 -37.55625 \n", "L 354.277864 -37.55625 \n", "L 353.858841 -37.55625 \n", "L 353.439817 -37.55625 \n", "L 353.020794 -37.55625 \n", "L 352.60177 -37.55625 \n", "L 352.182747 -37.55625 \n", "L 351.763724 -37.55625 \n", "L 351.3447 -37.55625 \n", "L 350.925677 -37.55625 \n", "L 350.506654 -37.55625 \n", "L 350.08763 -37.55625 \n", "L 349.668607 -37.55625 \n", "L 349.249583 -37.55625 \n", "L 348.83056 -37.55625 \n", "L 348.411537 -37.55625 \n", "L 347.992513 -37.55625 \n", "L 347.57349 -37.55625 \n", "L 347.154467 -37.55625 \n", "L 346.735443 -37.55625 \n", "L 346.31642 -37.55625 \n", "L 345.897396 -37.55625 \n", "L 345.478373 -37.55625 \n", "L 345.05934 -37.55625 \n", "L 344.640316 -37.55625 \n", "L 344.221293 -37.55625 \n", "L 343.80227 -37.55625 \n", "L 343.383246 -37.55625 \n", "L 342.964223 -37.55625 \n", "L 342.545199 -37.55625 \n", "L 342.126176 -37.55625 \n", "L 341.707153 -37.55625 \n", "L 341.288129 -37.55625 \n", "L 340.869106 -37.55625 \n", "L 340.450082 -37.55625 \n", "L 340.031059 -37.55625 \n", "L 339.612036 -37.55625 \n", "L 339.193012 -37.55625 \n", "L 338.773989 -37.55625 \n", "L 338.354966 -37.55625 \n", "L 337.935942 -37.55625 \n", "L 337.516919 -37.55625 \n", "L 337.097895 -37.55625 \n", "L 336.678872 -37.55625 \n", "L 336.259849 -37.55625 \n", "L 335.840825 -37.55625 \n", "L 335.421802 -37.55625 \n", "L 335.002779 -37.55625 \n", "L 334.583755 -37.55625 \n", "L 334.164732 -37.55625 \n", "L 333.745708 -37.55625 \n", "L 333.326685 -37.55625 \n", "L 332.907662 -37.55625 \n", "L 332.488638 -37.55625 \n", "L 332.069615 -37.55625 \n", "L 331.650582 -37.55625 \n", "L 331.231558 -37.55625 \n", "L 330.812535 -37.55625 \n", "L 330.393511 -37.55625 \n", "L 329.974488 -37.55625 \n", "L 329.555465 -37.55625 \n", "L 329.136441 -37.55625 \n", "L 328.717418 -37.55625 \n", "L 328.298394 -37.55625 \n", "L 327.879371 -37.55625 \n", "L 327.460348 -37.55625 \n", "L 327.041324 -37.55625 \n", "L 326.622301 -37.55625 \n", "L 326.203278 -37.55625 \n", "L 325.784254 -37.55625 \n", "L 325.365231 -37.55625 \n", "L 324.946207 -37.55625 \n", "L 324.527184 -37.55625 \n", "L 324.108161 -37.55625 \n", "L 323.689137 -37.55625 \n", "L 323.270114 -37.55625 \n", "L 322.851091 -37.55625 \n", "L 322.432067 -37.55625 \n", "L 322.013044 -37.55625 \n", "L 321.59401 -37.55625 \n", "L 321.174987 -37.55625 \n", "L 320.755964 -37.55625 \n", "L 320.33694 -37.55625 \n", "L 319.917917 -37.55625 \n", "L 319.498894 -37.55625 \n", "L 319.07987 -37.55625 \n", "L 318.660847 -37.55625 \n", "L 318.241823 -37.55625 \n", "L 317.8228 -37.55625 \n", "L 317.403777 -37.55625 \n", "L 316.984753 -37.55625 \n", "L 316.56573 -37.55625 \n", "L 316.146706 -37.55625 \n", "L 315.727683 -37.55625 \n", "L 315.30866 -37.55625 \n", "L 314.889636 -37.55625 \n", "L 314.470613 -37.55625 \n", "L 314.05159 -37.55625 \n", "L 313.632566 -37.55625 \n", "L 313.213543 -37.55625 \n", "L 312.794519 -37.55625 \n", "L 312.375496 -37.55625 \n", "L 311.956473 -37.55625 \n", "L 311.537439 -37.55625 \n", "L 311.118416 -37.55625 \n", "L 310.699393 -37.55625 \n", "L 310.280369 -37.55625 \n", "L 309.861346 -37.55625 \n", "L 309.442322 -37.55625 \n", "L 309.023299 -37.55625 \n", "L 308.604276 -37.55625 \n", "L 308.185252 -37.55625 \n", "L 307.766229 -37.55625 \n", "L 307.347206 -37.55625 \n", "L 306.928182 -37.55625 \n", "L 306.509159 -37.55625 \n", "L 306.090135 -37.55625 \n", "L 305.671112 -37.55625 \n", "L 305.252089 -37.55625 \n", "L 304.833065 -37.55625 \n", "L 304.414042 -37.55625 \n", "L 303.995018 -37.55625 \n", "L 303.575995 -37.55625 \n", "L 303.156972 -37.55625 \n", "L 302.737948 -37.55625 \n", "L 302.318925 -37.55625 \n", "L 301.899902 -37.55625 \n", "L 301.480868 -37.55625 \n", "L 301.061845 -37.55625 \n", "L 300.642821 -37.55625 \n", "L 300.223798 -37.55625 \n", "L 299.804775 -37.55625 \n", "L 299.385751 -37.55625 \n", "L 298.966728 -37.55625 \n", "L 298.547705 -37.55625 \n", "L 298.128681 -37.55625 \n", "L 297.709658 -37.55625 \n", "L 297.290634 -37.55625 \n", "L 296.871611 -37.55625 \n", "L 296.452588 -37.55625 \n", "L 296.033564 -37.55625 \n", "L 295.614541 -37.55625 \n", "L 295.195518 -37.55625 \n", "L 294.776494 -37.55625 \n", "L 294.357466 -37.55625 \n", "L 293.938442 -37.55625 \n", "L 293.519419 -37.55625 \n", "L 293.100396 -37.55625 \n", "z\n", "\" style=\"stroke: #7f7f7f; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pc8b7a1e7e3)\">\n", "     <use xlink:href=\"#m32cb1aa068\" x=\"0\" y=\"226.194375\" style=\"fill: #7f7f7f; fill-opacity: 0.5; stroke: #7f7f7f; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"line2d_1\">\n", "    <path d=\"M 43.78125 188.638125 \n", "L 43.78125 177.955718 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #1f77b4; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_2\">\n", "    <path d=\"M 129.681123 188.638125 \n", "L 129.681123 122.69085 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #1f77b4; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_3\">\n", "    <path d=\"M 130.100146 188.638125 \n", "L 130.100146 122.194516 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #ff7f0e; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_4\">\n", "    <path d=\"M 165.298145 188.638125 \n", "L 165.298145 75.316811 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #ff7f0e; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_5\">\n", "    <path d=\"M 165.717168 188.638125 \n", "L 165.717168 74.750347 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #2ca02c; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_6\">\n", "    <path d=\"M 189.601526 188.638125 \n", "L 189.601526 47.215647 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #2ca02c; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_7\">\n", "    <path d=\"M 190.020549 188.638125 \n", "L 190.020549 46.85888 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #d62728; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_8\">\n", "    <path d=\"M 210.971738 188.638125 \n", "L 210.971738 37.44191 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #d62728; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_9\">\n", "    <path d=\"M 211.390762 188.638125 \n", "L 211.390762 37.438125 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #9467bd; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_10\">\n", "    <path d=\"M 232.760974 188.638125 \n", "L 232.760974 46.85888 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #9467bd; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_11\">\n", "    <path d=\"M 233.179997 188.638125 \n", "L 233.179997 47.215665 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #8c564b; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_12\">\n", "    <path d=\"M 257.06435 188.638125 \n", "L 257.06435 74.750338 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #8c564b; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_13\">\n", "    <path d=\"M 257.483373 188.638125 \n", "L 257.483373 75.316811 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #e377c2; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_14\">\n", "    <path d=\"M 292.681372 188.638125 \n", "L 292.681372 122.194502 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #e377c2; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_15\">\n", "    <path d=\"M 293.100396 188.638125 \n", "L 293.100396 122.690832 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #7f7f7f; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_16\">\n", "    <path d=\"M 378.58125 188.638125 \n", "L 378.58125 177.85226 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #7f7f7f; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_17\">\n", "    <path d=\"M 43.78125 177.955718 \n", "L 50.066601 176.29888 \n", "L 55.932938 174.533085 \n", "L 61.799275 172.529344 \n", "L 67.246579 170.431691 \n", "L 72.27487 168.272087 \n", "L 77.30315 165.878427 \n", "L 82.331441 163.231008 \n", "L 86.940698 160.564336 \n", "L 91.549955 157.652606 \n", "L 96.159222 154.481293 \n", "L 100.76848 151.036869 \n", "L 105.377747 147.307362 \n", "L 109.987004 143.283057 \n", "L 114.596271 138.957195 \n", "L 119.205528 134.326878 \n", "L 123.814785 129.393975 \n", "L 128.843076 123.676666 \n", "L 133.871367 117.627676 \n", "L 139.31867 110.736061 \n", "L 145.604026 102.419105 \n", "L 153.984504 90.923591 \n", "L 167.393267 72.499923 \n", "L 172.840576 65.405944 \n", "L 177.449835 59.758912 \n", "L 181.221048 55.457045 \n", "L 184.992264 51.501572 \n", "L 188.344453 48.319897 \n", "L 191.277621 45.823403 \n", "L 194.210787 43.618437 \n", "L 196.72493 41.975707 \n", "L 199.239072 40.573072 \n", "L 201.753215 39.420237 \n", "L 204.267357 38.525305 \n", "L 206.7815 37.894567 \n", "L 209.295643 37.5326 \n", "L 211.390762 37.438125 \n", "L 213.485881 37.532582 \n", "L 215.581 37.815503 \n", "L 218.095142 38.401676 \n", "L 220.609285 39.252988 \n", "L 223.123428 40.363339 \n", "L 225.63757 41.724978 \n", "L 228.151713 43.328423 \n", "L 231.084879 45.490077 \n", "L 234.018044 47.946232 \n", "L 236.951213 50.675079 \n", "L 240.303402 54.097353 \n", "L 244.074618 58.289893 \n", "L 248.264854 63.3087 \n", "L 252.874114 69.178802 \n", "L 258.740449 77.024438 \n", "L 268.79702 90.923591 \n", "L 278.434567 104.108995 \n", "L 284.719923 112.355179 \n", "L 290.586255 119.679209 \n", "L 295.614541 125.620445 \n", "L 300.642821 131.222419 \n", "L 305.252089 136.045855 \n", "L 309.861346 140.56539 \n", "L 314.470613 144.781046 \n", "L 319.07987 148.69719 \n", "L 323.689137 152.32178 \n", "L 328.298394 155.665416 \n", "L 332.907662 158.740704 \n", "L 337.516919 161.561602 \n", "L 342.545199 164.366214 \n", "L 347.57349 166.905489 \n", "L 352.60177 169.199253 \n", "L 358.049084 171.429737 \n", "L 363.496388 173.418945 \n", "L 369.362726 175.317454 \n", "L 375.229053 176.989208 \n", "L 378.58125 177.85226 \n", "L 378.58125 177.85226 \n", "\" clip-path=\"url(#pc8b7a1e7e3)\" style=\"fill: none; stroke: #000000; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"patch_3\">\n", "    <path d=\"M 43.78125 188.638125 \n", "L 43.78125 22.318125 \n", "\" style=\"fill: none; stroke: #262626; stroke-width: 0.8; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_4\">\n", "    <path d=\"M 378.58125 188.638125 \n", "L 378.58125 22.318125 \n", "\" style=\"fill: none; stroke: #262626; stroke-width: 0.8; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_5\">\n", "    <path d=\"M 43.78125 188.638125 \n", "L 378.58125 188.638125 \n", "\" style=\"fill: none; stroke: #262626; stroke-width: 0.8; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_6\">\n", "    <path d=\"M 43.78125 22.318125 \n", "L 378.58125 22.318125 \n", "\" style=\"fill: none; stroke: #262626; stroke-width: 0.8; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"text_17\">\n", "    <!-- Dequantization distribution for 8 discrete values -->\n", "    <g style=\"fill: #262626\" transform=\"translate(66.164062 16.318125)scale(0.12 -0.12)\">\n", "     <defs>\n", "      <path id=\"DejaVuSans-44\" d=\"M 1259 4147 \n", "L 1259 519 \n", "L 2022 519 \n", "Q 2988 519 3436 956 \n", "Q 3884 1394 3884 2338 \n", "Q 3884 3275 3436 3711 \n", "Q 2988 4147 2022 4147 \n", "L 1259 4147 \n", "z\n", "M 628 4666 \n", "L 1925 4666 \n", "Q 3281 4666 3915 4102 \n", "Q 4550 3538 4550 2338 \n", "Q 4550 1131 3912 565 \n", "Q 3275 0 1925 0 \n", "L 628 0 \n", "L 628 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-65\" d=\"M 3597 1894 \n", "L 3597 1613 \n", "L 953 1613 \n", "Q 991 1019 1311 708 \n", "Q 1631 397 2203 397 \n", "Q 2534 397 2845 478 \n", "Q 3156 559 3463 722 \n", "L 3463 178 \n", "Q 3153 47 2828 -22 \n", "Q 2503 -91 2169 -91 \n", "Q 1331 -91 842 396 \n", "Q 353 884 353 1716 \n", "Q 353 2575 817 3079 \n", "Q 1281 3584 2069 3584 \n", "Q 2775 3584 3186 3129 \n", "Q 3597 2675 3597 1894 \n", "z\n", "M 3022 2063 \n", "Q 3016 2534 2758 2815 \n", "Q 2500 3097 2075 3097 \n", "Q 1594 3097 1305 2825 \n", "Q 1016 2553 972 2059 \n", "L 3022 2063 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-71\" d=\"M 947 1747 \n", "Q 947 1113 1208 752 \n", "Q 1469 391 1925 391 \n", "Q 2381 391 2643 752 \n", "Q 2906 1113 2906 1747 \n", "Q 2906 2381 2643 2742 \n", "Q 2381 3103 1925 3103 \n", "Q 1469 3103 1208 2742 \n", "Q 947 2381 947 1747 \n", "z\n", "M 2906 525 \n", "Q 2725 213 2448 61 \n", "Q 2172 -91 1784 -91 \n", "Q 1150 -91 751 415 \n", "Q 353 922 353 1747 \n", "Q 353 2572 751 3078 \n", "Q 1150 3584 1784 3584 \n", "Q 2172 3584 2448 3432 \n", "Q 2725 3281 2906 2969 \n", "L 2906 3500 \n", "L 3481 3500 \n", "L 3481 -1331 \n", "L 2906 -1331 \n", "L 2906 525 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-75\" d=\"M 544 1381 \n", "L 544 3500 \n", "L 1119 3500 \n", "L 1119 1403 \n", "Q 1119 906 1312 657 \n", "Q 1506 409 1894 409 \n", "Q 2359 409 2629 706 \n", "Q 2900 1003 2900 1516 \n", "L 2900 3500 \n", "L 3475 3500 \n", "L 3475 0 \n", "L 2900 0 \n", "L 2900 538 \n", "Q 2691 219 2414 64 \n", "Q 2138 -91 1772 -91 \n", "Q 1169 -91 856 284 \n", "Q 544 659 544 1381 \n", "z\n", "M 1991 3584 \n", "L 1991 3584 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6e\" d=\"M 3513 2113 \n", "L 3513 0 \n", "L 2938 0 \n", "L 2938 2094 \n", "Q 2938 2591 2744 2837 \n", "Q 2550 3084 2163 3084 \n", "Q 1697 3084 1428 2787 \n", "Q 1159 2491 1159 1978 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 3500 \n", "L 1159 3500 \n", "L 1159 2956 \n", "Q 1366 3272 1645 3428 \n", "Q 1925 3584 2291 3584 \n", "Q 2894 3584 3203 3211 \n", "Q 3513 2838 3513 2113 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-20\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-64\" d=\"M 2906 2969 \n", "L 2906 4863 \n", "L 3481 4863 \n", "L 3481 0 \n", "L 2906 0 \n", "L 2906 525 \n", "Q 2725 213 2448 61 \n", "Q 2172 -91 1784 -91 \n", "Q 1150 -91 751 415 \n", "Q 353 922 353 1747 \n", "Q 353 2572 751 3078 \n", "Q 1150 3584 1784 3584 \n", "Q 2172 3584 2448 3432 \n", "Q 2725 3281 2906 2969 \n", "z\n", "M 947 1747 \n", "Q 947 1113 1208 752 \n", "Q 1469 391 1925 391 \n", "Q 2381 391 2643 752 \n", "Q 2906 1113 2906 1747 \n", "Q 2906 2381 2643 2742 \n", "Q 2381 3103 1925 3103 \n", "Q 1469 3103 1208 2742 \n", "Q 947 2381 947 1747 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-73\" d=\"M 2834 3397 \n", "L 2834 2853 \n", "Q 2591 2978 2328 3040 \n", "Q 2066 3103 1784 3103 \n", "Q 1356 3103 1142 2972 \n", "Q 928 2841 928 2578 \n", "Q 928 2378 1081 2264 \n", "Q 1234 2150 1697 2047 \n", "L 1894 2003 \n", "Q 2506 1872 2764 1633 \n", "Q 3022 1394 3022 966 \n", "Q 3022 478 2636 193 \n", "Q 2250 -91 1575 -91 \n", "Q 1294 -91 989 -36 \n", "Q 684 19 347 128 \n", "L 347 722 \n", "Q 666 556 975 473 \n", "Q 1284 391 1588 391 \n", "Q 1994 391 2212 530 \n", "Q 2431 669 2431 922 \n", "Q 2431 1156 2273 1281 \n", "Q 2116 1406 1581 1522 \n", "L 1381 1569 \n", "Q 847 1681 609 1914 \n", "Q 372 2147 372 2553 \n", "Q 372 3047 722 3315 \n", "Q 1072 3584 1716 3584 \n", "Q 2034 3584 2315 3537 \n", "Q 2597 3491 2834 3397 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-66\" d=\"M 2375 4863 \n", "L 2375 4384 \n", "L 1825 4384 \n", "Q 1516 4384 1395 4259 \n", "Q 1275 4134 1275 3809 \n", "L 1275 3500 \n", "L 2222 3500 \n", "L 2222 3053 \n", "L 1275 3053 \n", "L 1275 0 \n", "L 697 0 \n", "L 697 3053 \n", "L 147 3053 \n", "L 147 3500 \n", "L 697 3500 \n", "L 697 3744 \n", "Q 697 4328 969 4595 \n", "Q 1241 4863 1831 4863 \n", "L 2375 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-38\" d=\"M 2034 2216 \n", "Q 1584 2216 1326 1975 \n", "Q 1069 1734 1069 1313 \n", "Q 1069 891 1326 650 \n", "Q 1584 409 2034 409 \n", "Q 2484 409 2743 651 \n", "Q 3003 894 3003 1313 \n", "Q 3003 1734 2745 1975 \n", "Q 2488 2216 2034 2216 \n", "z\n", "M 1403 2484 \n", "Q 997 2584 770 2862 \n", "Q 544 3141 544 3541 \n", "Q 544 4100 942 4425 \n", "Q 1341 4750 2034 4750 \n", "Q 2731 4750 3128 4425 \n", "Q 3525 4100 3525 3541 \n", "Q 3525 3141 3298 2862 \n", "Q 3072 2584 2669 2484 \n", "Q 3125 2378 3379 2068 \n", "Q 3634 1759 3634 1313 \n", "Q 3634 634 3220 271 \n", "Q 2806 -91 2034 -91 \n", "Q 1263 -91 848 271 \n", "Q 434 634 434 1313 \n", "Q 434 1759 690 2068 \n", "Q 947 2378 1403 2484 \n", "z\n", "M 1172 3481 \n", "Q 1172 3119 1398 2916 \n", "Q 1625 2713 2034 2713 \n", "Q 2441 2713 2670 2916 \n", "Q 2900 3119 2900 3481 \n", "Q 2900 3844 2670 4047 \n", "Q 2441 4250 2034 4250 \n", "Q 1625 4250 1398 4047 \n", "Q 1172 3844 1172 3481 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-63\" d=\"M 3122 3366 \n", "L 3122 2828 \n", "Q 2878 2963 2633 3030 \n", "Q 2388 3097 2138 3097 \n", "Q 1578 3097 1268 2742 \n", "Q 959 2388 959 1747 \n", "Q 959 1106 1268 751 \n", "Q 1578 397 2138 397 \n", "Q 2388 397 2633 464 \n", "Q 2878 531 3122 666 \n", "L 3122 134 \n", "Q 2881 22 2623 -34 \n", "Q 2366 -91 2075 -91 \n", "Q 1284 -91 818 406 \n", "Q 353 903 353 1747 \n", "Q 353 2603 823 3093 \n", "Q 1294 3584 2113 3584 \n", "Q 2378 3584 2631 3529 \n", "Q 2884 3475 3122 3366 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-76\" d=\"M 191 3500 \n", "L 800 3500 \n", "L 1894 563 \n", "L 2988 3500 \n", "L 3597 3500 \n", "L 2284 0 \n", "L 1503 0 \n", "L 191 3500 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "     </defs>\n", "     <use xlink:href=\"#DejaVuSans-44\"/>\n", "     <use xlink:href=\"#DejaVuSans-65\" x=\"77.001953\"/>\n", "     <use xlink:href=\"#DejaVuSans-71\" x=\"138.525391\"/>\n", "     <use xlink:href=\"#DejaVuSans-75\" x=\"202.001953\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"265.380859\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"326.660156\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"390.039062\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"429.248047\"/>\n", "     <use xlink:href=\"#DejaVuSans-7a\" x=\"457.03125\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"509.521484\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"570.800781\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"610.009766\"/>\n", "     <use xlink:href=\"#DejaVuSans-6f\" x=\"637.792969\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"698.974609\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"762.353516\"/>\n", "     <use xlink:href=\"#DejaVuSans-64\" x=\"794.140625\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"857.617188\"/>\n", "     <use xlink:href=\"#DejaVuSans-73\" x=\"885.400391\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"937.5\"/>\n", "     <use xlink:href=\"#DejaVuSans-72\" x=\"976.708984\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"1017.822266\"/>\n", "     <use xlink:href=\"#DejaVuSans-62\" x=\"1045.605469\"/>\n", "     <use xlink:href=\"#DejaVuSans-75\" x=\"1109.082031\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"1172.460938\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"1211.669922\"/>\n", "     <use xlink:href=\"#DejaVuSans-6f\" x=\"1239.453125\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"1300.634766\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"1364.013672\"/>\n", "     <use xlink:href=\"#DejaVuSans-66\" x=\"1395.800781\"/>\n", "     <use xlink:href=\"#DejaVuSans-6f\" x=\"1431.005859\"/>\n", "     <use xlink:href=\"#DejaVuSans-72\" x=\"1492.1875\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"1533.300781\"/>\n", "     <use xlink:href=\"#DejaVuSans-38\" x=\"1565.087891\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"1628.710938\"/>\n", "     <use xlink:href=\"#DejaVuSans-64\" x=\"1660.498047\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"1723.974609\"/>\n", "     <use xlink:href=\"#DejaVuSans-73\" x=\"1751.757812\"/>\n", "     <use xlink:href=\"#DejaVuSans-63\" x=\"1803.857422\"/>\n", "     <use xlink:href=\"#DejaVuSans-72\" x=\"1858.837891\"/>\n", "     <use xlink:href=\"#DejaVuSans-65\" x=\"1897.701172\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"1959.224609\"/>\n", "     <use xlink:href=\"#DejaVuSans-65\" x=\"1998.433594\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"2059.957031\"/>\n", "     <use xlink:href=\"#DejaVuSans-76\" x=\"2091.744141\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"2150.923828\"/>\n", "     <use xlink:href=\"#DejaVuSans-6c\" x=\"2212.203125\"/>\n", "     <use xlink:href=\"#DejaVuSans-75\" x=\"2239.986328\"/>\n", "     <use xlink:href=\"#DejaVuSans-65\" x=\"2303.365234\"/>\n", "     <use xlink:href=\"#DejaVuSans-73\" x=\"2364.888672\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"legend_1\">\n", "    <g id=\"patch_7\">\n", "     <path d=\"M 333.21875 147.743125 \n", "L 371.58125 147.743125 \n", "Q 373.58125 147.743125 373.58125 145.743125 \n", "L 373.58125 29.318125 \n", "Q 373.58125 27.318125 371.58125 27.318125 \n", "L 333.21875 27.318125 \n", "Q 331.21875 27.318125 331.21875 29.318125 \n", "L 331.21875 145.743125 \n", "Q 331.21875 147.743125 333.21875 147.743125 \n", "z\n", "\" style=\"fill: #ffffff; opacity: 0.8; stroke: #cccccc; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"patch_8\">\n", "     <path d=\"M 335.21875 38.916562 \n", "L 355.21875 38.916562 \n", "L 355.21875 31.916562 \n", "L 335.21875 31.916562 \n", "z\n", "\" style=\"fill: #1f77b4; fill-opacity: 0.5; stroke: #1f77b4; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_18\">\n", "     <!-- 0 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(363.21875 38.916562)scale(0.1 -0.1)\">\n", "      <use xlink:href=\"#DejaVuSans-30\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_9\">\n", "     <path d=\"M 335.21875 53.594688 \n", "L 355.21875 53.594688 \n", "L 355.21875 46.594688 \n", "L 335.21875 46.594688 \n", "z\n", "\" style=\"fill: #ff7f0e; fill-opacity: 0.5; stroke: #ff7f0e; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_19\">\n", "     <!-- 1 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(363.21875 53.594688)scale(0.1 -0.1)\">\n", "      <use xlink:href=\"#DejaVuSans-31\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_10\">\n", "     <path d=\"M 335.21875 68.272812 \n", "L 355.21875 68.272812 \n", "L 355.21875 61.272812 \n", "L 335.21875 61.272812 \n", "z\n", "\" style=\"fill: #2ca02c; fill-opacity: 0.5; stroke: #2ca02c; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_20\">\n", "     <!-- 2 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(363.21875 68.272812)scale(0.1 -0.1)\">\n", "      <use xlink:href=\"#DejaVuSans-32\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_11\">\n", "     <path d=\"M 335.21875 82.950938 \n", "L 355.21875 82.950938 \n", "L 355.21875 75.950938 \n", "L 335.21875 75.950938 \n", "z\n", "\" style=\"fill: #d62728; fill-opacity: 0.5; stroke: #d62728; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_21\">\n", "     <!-- 3 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(363.21875 82.950938)scale(0.1 -0.1)\">\n", "      <defs>\n", "       <path id=\"DejaVuSans-33\" d=\"M 2597 2516 \n", "Q 3050 2419 3304 2112 \n", "Q 3559 1806 3559 1356 \n", "Q 3559 666 3084 287 \n", "Q 2609 -91 1734 -91 \n", "Q 1441 -91 1130 -33 \n", "Q 819 25 488 141 \n", "L 488 750 \n", "Q 750 597 1062 519 \n", "Q 1375 441 1716 441 \n", "Q 2309 441 2620 675 \n", "Q 2931 909 2931 1356 \n", "Q 2931 1769 2642 2001 \n", "Q 2353 2234 1838 2234 \n", "L 1294 2234 \n", "L 1294 2753 \n", "L 1863 2753 \n", "Q 2328 2753 2575 2939 \n", "Q 2822 3125 2822 3475 \n", "Q 2822 3834 2567 4026 \n", "Q 2313 4219 1838 4219 \n", "Q 1578 4219 1281 4162 \n", "Q 984 4106 628 3988 \n", "L 628 4550 \n", "Q 988 4650 1302 4700 \n", "Q 1616 4750 1894 4750 \n", "Q 2613 4750 3031 4423 \n", "Q 3450 4097 3450 3541 \n", "Q 3450 3153 3228 2886 \n", "Q 3006 2619 2597 2516 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      </defs>\n", "      <use xlink:href=\"#DejaVuSans-33\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_12\">\n", "     <path d=\"M 335.21875 97.629063 \n", "L 355.21875 97.629063 \n", "L 355.21875 90.629063 \n", "L 335.21875 90.629063 \n", "z\n", "\" style=\"fill: #9467bd; fill-opacity: 0.5; stroke: #9467bd; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_22\">\n", "     <!-- 4 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(363.21875 97.629063)scale(0.1 -0.1)\">\n", "      <use xlink:href=\"#DejaVuSans-34\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_13\">\n", "     <path d=\"M 335.21875 112.307187 \n", "L 355.21875 112.307187 \n", "L 355.21875 105.307187 \n", "L 335.21875 105.307187 \n", "z\n", "\" style=\"fill: #8c564b; fill-opacity: 0.5; stroke: #8c564b; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_23\">\n", "     <!-- 5 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(363.21875 112.307187)scale(0.1 -0.1)\">\n", "      <use xlink:href=\"#DejaVuSans-35\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_14\">\n", "     <path d=\"M 335.21875 126.985312 \n", "L 355.21875 126.985312 \n", "L 355.21875 119.985312 \n", "L 335.21875 119.985312 \n", "z\n", "\" style=\"fill: #e377c2; fill-opacity: 0.5; stroke: #e377c2; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_24\">\n", "     <!-- 6 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(363.21875 126.985312)scale(0.1 -0.1)\">\n", "      <defs>\n", "       <path id=\"DejaVuSans-36\" d=\"M 2113 2584 \n", "Q 1688 2584 1439 2293 \n", "Q 1191 2003 1191 1497 \n", "Q 1191 994 1439 701 \n", "Q 1688 409 2113 409 \n", "Q 2538 409 2786 701 \n", "Q 3034 994 3034 1497 \n", "Q 3034 2003 2786 2293 \n", "Q 2538 2584 2113 2584 \n", "z\n", "M 3366 4563 \n", "L 3366 3988 \n", "Q 3128 4100 2886 4159 \n", "Q 2644 4219 2406 4219 \n", "Q 1781 4219 1451 3797 \n", "Q 1122 3375 1075 2522 \n", "Q 1259 2794 1537 2939 \n", "Q 1816 3084 2150 3084 \n", "Q 2853 3084 3261 2657 \n", "Q 3669 2231 3669 1497 \n", "Q 3669 778 3244 343 \n", "Q 2819 -91 2113 -91 \n", "Q 1303 -91 875 529 \n", "Q 447 1150 447 2328 \n", "Q 447 3434 972 4092 \n", "Q 1497 4750 2381 4750 \n", "Q 2619 4750 2861 4703 \n", "Q 3103 4656 3366 4563 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      </defs>\n", "      <use xlink:href=\"#DejaVuSans-36\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_15\">\n", "     <path d=\"M 335.21875 141.663437 \n", "L 355.21875 141.663437 \n", "L 355.21875 134.663437 \n", "L 335.21875 134.663437 \n", "z\n", "\" style=\"fill: #7f7f7f; fill-opacity: 0.5; stroke: #7f7f7f; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_25\">\n", "     <!-- 7 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(363.21875 141.663437)scale(0.1 -0.1)\">\n", "      <defs>\n", "       <path id=\"DejaVuSans-37\" d=\"M 525 4666 \n", "L 3525 4666 \n", "L 3525 4397 \n", "L 1831 0 \n", "L 1172 0 \n", "L 2766 4134 \n", "L 525 4134 \n", "L 525 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      </defs>\n", "      <use xlink:href=\"#DejaVuSans-37\"/>\n", "     </g>\n", "    </g>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"pc8b7a1e7e3\">\n", "   <rect x=\"43.78125\" y=\"22.318125\" width=\"334.8\" height=\"166.32\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 600x300 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["\n", "\n", "def visualize_dequantization(quants, prior=None):\n", "    \"\"\"Function for visualizing the dequantization values of discrete values in continuous space.\"\"\"\n", "    # Prior over discrete values. If not given, a uniform is assumed\n", "    if prior is None:\n", "        prior = np.ones(quants, dtype=np.float32) / quants\n", "    prior = prior / prior.sum() * quants  # In the following, we assume 1 for each value means uniform distribution\n", "\n", "    inp = torch.arange(-4, 4, 0.01).view(-1, 1, 1, 1)  # Possible continuous values we want to consider\n", "    ldj = torch.zeros(inp.shape[0])\n", "    dequant_module = Dequantization(quants=quants)\n", "    # Invert dequantization on continuous values to find corresponding discrete value\n", "    out, ldj = dequant_module.forward(inp, ldj, reverse=True)\n", "    inp, out, prob = inp.squeeze().numpy(), out.squeeze().numpy(), ldj.exp().numpy()\n", "    prob = prob * prior[out]  # Probability scaled by categorical prior\n", "\n", "    # Plot volumes and continuous distribution\n", "    sns.set_style(\"white\")\n", "    _ = plt.figure(figsize=(6, 3))\n", "    x_ticks = []\n", "    for v in np.unique(out):\n", "        indices = np.where(out == v)\n", "        color = to_rgb(\"C%i\" % v)\n", "        plt.fill_between(inp[indices], prob[indices], np.zeros(indices[0].shape[0]), color=color + (0.5,), label=str(v))\n", "        plt.plot([inp[indices[0][0]]] * 2, [0, prob[indices[0][0]]], color=color)\n", "        plt.plot([inp[indices[0][-1]]] * 2, [0, prob[indices[0][-1]]], color=color)\n", "        x_ticks.append(inp[indices[0][0]])\n", "    x_ticks.append(inp.max())\n", "    plt.xticks(x_ticks, [\"%.1f\" % x for x in x_ticks])\n", "    plt.plot(inp, prob, color=(0.0, 0.0, 0.0))\n", "    # Set final plot properties\n", "    plt.ylim(0, prob.max() * 1.1)\n", "    plt.xlim(inp.min(), inp.max())\n", "    plt.xlabel(\"z\")\n", "    plt.ylabel(\"Probability\")\n", "    plt.title(\"Dequantization distribution for %i discrete values\" % quants)\n", "    plt.legend()\n", "    plt.show()\n", "    plt.close()\n", "\n", "\n", "visualize_dequantization(quants=8)"]}, {"cell_type": "markdown", "id": "c276d1a9", "metadata": {"papermill": {"duration": 0.013705, "end_time": "2022-05-12T12:19:09.808110", "exception": false, "start_time": "2022-05-12T12:19:09.794405", "status": "completed"}, "tags": []}, "source": ["The visualized distribution show the sub-volumes that are assigned to the different discrete values.\n", "The value $0$ has its volume between $[-\\infty, -1.9)$, the value $1$ is represented by the interval $[-1.9, -1.1)$, etc.\n", "The volume for each discrete value has the same probability mass.\n", "That's why the volumes close to the center (e.g. 3 and 4) have a smaller area on the z-axis as others\n", "($z$ is being used to denote the output of the whole dequantization flow).\n", "\n", "Effectively, the consecutive normalizing flow models discrete images by the following objective:\n", "\n", "$$\\log p(x) = \\log \\mathbb{E}_{u\\sim q(u|x)}\\left[\\frac{p(x+u)}{q(u|x)} \\right] \\geq \\mathbb{E}_{u}\\left[\\log \\frac{p(x+u)}{q(u|x)} \\right]$$\n", "\n", "Although normalizing flows are exact in likelihood, we have a lower bound.\n", "Specifically, this is an example of the Jensen inequality because we need to move the log into the expectation so we can use Monte-carlo estimates.\n", "In general, this bound is considerably smaller than the ELBO in variational autoencoders.\n", "Actually, we can reduce the bound ourselves by estimating the expectation not by one, but by $M$ samples.\n", "In other words, we can apply importance sampling which leads to the following inequality:\n", "\n", "$$\\log p(x) = \\log \\mathbb{E}_{u\\sim q(u|x)}\\left[\\frac{p(x+u)}{q(u|x)} \\right] \\geq \\mathbb{E}_{u}\\left[\\log \\frac{1}{M} \\sum_{m=1}^{M} \\frac{p(x+u_m)}{q(u_m|x)} \\right] \\geq \\mathbb{E}_{u}\\left[\\log \\frac{p(x+u)}{q(u|x)} \\right]$$\n", "\n", "The importance sampling $\\frac{1}{M} \\sum_{m=1}^{M} \\frac{p(x+u_m)}{q(u_m|x)}$ becomes\n", "$\\mathbb{E}_{u\\sim q(u|x)}\\left[\\frac{p(x+u)}{q(u|x)} \\right]$ if $M\\to \\infty$,\n", "so that the more samples we use, the tighter the bound is.\n", "During testing, we can make use of this property and have it implemented in `test_step` in `ImageFlow`.\n", "In theory, we could also use this tighter bound during training.\n", "However, related work has shown that this does not necessarily lead to\n", "an improvement given the additional computational cost, and it is more\n", "efficient to stick with a single estimate [5]."]}, {"cell_type": "markdown", "id": "c38cf9bb", "metadata": {"papermill": {"duration": 0.013767, "end_time": "2022-05-12T12:19:09.837449", "exception": false, "start_time": "2022-05-12T12:19:09.823682", "status": "completed"}, "tags": []}, "source": ["### Variational Dequantization\n", "\n", "Dequantization uses a uniform distribution for the noise $u$ which effectively leads to images being represented as hypercubes\n", "(cube in high dimensions) with sharp borders.\n", "However, modeling such sharp borders is not easy for a flow as it uses smooth transformations to convert it into a Gaussian distribution.\n", "\n", "Another way of looking at it is if we change the prior distribution in the previous visualization.\n", "Imagine we have independent Gaussian noise on pixels which is commonly the case for any real-world taken picture.\n", "Therefore, the flow would have to model a distribution as above, but with the individual volumes scaled as follows:"]}, {"cell_type": "code", "execution_count": 10, "id": "20e5f49a", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:09.866816Z", "iopub.status.busy": "2022-05-12T12:19:09.866355Z", "iopub.status.idle": "2022-05-12T12:19:10.297455Z", "shell.execute_reply": "2022-05-12T12:19:10.296698Z"}, "papermill": {"duration": 0.448967, "end_time": "2022-05-12T12:19:10.300018", "exception": false, "start_time": "2022-05-12T12:19:09.851051", "status": "completed"}, "tags": []}, "outputs": [{"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"384.192188pt\" height=\"226.194375pt\" viewBox=\"0 0 384.192188 226.194375\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:10.029130</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 226.194375 \n", "L 384.192188 226.194375 \n", "L 384.192188 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g id=\"patch_2\">\n", "    <path d=\"M 34.240625 188.638125 \n", "L 369.040625 188.638125 \n", "L 369.040625 22.318125 \n", "L 34.240625 22.318125 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "   </g>\n", "   <g id=\"matplotlib.axis_1\">\n", "    <g id=\"xtick_1\">\n", "     <g id=\"text_1\">\n", "      <!-- -4.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(24.485156 203.236563)scale(0.1 -0.1)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-2d\" d=\"M 313 2009 \n", "L 1997 2009 \n", "L 1997 1497 \n", "L 313 1497 \n", "L 313 2009 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-34\" d=\"M 2419 4116 \n", "L 825 1625 \n", "L 2419 1625 \n", "L 2419 4116 \n", "z\n", "M 2253 4666 \n", "L 3047 4666 \n", "L 3047 1625 \n", "L 3713 1625 \n", "L 3713 1100 \n", "L 3047 1100 \n", "L 3047 0 \n", "L 2419 0 \n", "L 2419 1100 \n", "L 313 1100 \n", "L 313 1709 \n", "L 2253 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-2e\" d=\"M 684 794 \n", "L 1344 794 \n", "L 1344 0 \n", "L 684 0 \n", "L 684 794 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-30\" d=\"M 2034 4250 \n", "Q 1547 4250 1301 3770 \n", "Q 1056 3291 1056 2328 \n", "Q 1056 1369 1301 889 \n", "Q 1547 409 2034 409 \n", "Q 2525 409 2770 889 \n", "Q 3016 1369 3016 2328 \n", "Q 3016 3291 2770 3770 \n", "Q 2525 4250 2034 4250 \n", "z\n", "M 2034 4750 \n", "Q 2819 4750 3233 4129 \n", "Q 3647 3509 3647 2328 \n", "Q 3647 1150 3233 529 \n", "Q 2819 -91 2034 -91 \n", "Q 1250 -91 836 529 \n", "Q 422 1150 422 2328 \n", "Q 422 3509 836 4129 \n", "Q 1250 4750 2034 4750 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2d\"/>\n", "       <use xlink:href=\"#DejaVuSans-34\" x=\"36.083984\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"99.707031\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"131.494141\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_2\">\n", "     <g id=\"text_2\">\n", "      <!-- -1.9 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(110.804052 203.236563)scale(0.1 -0.1)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-31\" d=\"M 794 531 \n", "L 1825 531 \n", "L 1825 4091 \n", "L 703 3866 \n", "L 703 4441 \n", "L 1819 4666 \n", "L 2450 4666 \n", "L 2450 531 \n", "L 3481 531 \n", "L 3481 0 \n", "L 794 0 \n", "L 794 531 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-39\" d=\"M 703 97 \n", "L 703 672 \n", "Q 941 559 1184 500 \n", "Q 1428 441 1663 441 \n", "Q 2288 441 2617 861 \n", "Q 2947 1281 2994 2138 \n", "Q 2813 1869 2534 1725 \n", "Q 2256 1581 1919 1581 \n", "Q 1219 1581 811 2004 \n", "Q 403 2428 403 3163 \n", "Q 403 3881 828 4315 \n", "Q 1253 4750 1959 4750 \n", "Q 2769 4750 3195 4129 \n", "Q 3622 3509 3622 2328 \n", "Q 3622 1225 3098 567 \n", "Q 2575 -91 1691 -91 \n", "Q 1453 -91 1209 -44 \n", "Q 966 3 703 97 \n", "z\n", "M 1959 2075 \n", "Q 2384 2075 2632 2365 \n", "Q 2881 2656 2881 3163 \n", "Q 2881 3666 2632 3958 \n", "Q 2384 4250 1959 4250 \n", "Q 1534 4250 1286 3958 \n", "Q 1038 3666 1038 3163 \n", "Q 1038 2656 1286 2365 \n", "Q 1534 2075 1959 2075 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2d\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"36.083984\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"99.707031\"/>\n", "       <use xlink:href=\"#DejaVuSans-39\" x=\"131.494141\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_3\">\n", "     <g id=\"text_3\">\n", "      <!-- -1.1 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(146.421075 203.236563)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-2d\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"36.083984\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"99.707031\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"131.494141\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_4\">\n", "     <g id=\"text_4\">\n", "      <!-- -0.5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(170.724456 203.236563)scale(0.1 -0.1)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-35\" d=\"M 691 4666 \n", "L 3169 4666 \n", "L 3169 4134 \n", "L 1269 4134 \n", "L 1269 2991 \n", "Q 1406 3038 1543 3061 \n", "Q 1681 3084 1819 3084 \n", "Q 2600 3084 3056 2656 \n", "Q 3513 2228 3513 1497 \n", "Q 3513 744 3044 326 \n", "Q 2575 -91 1722 -91 \n", "Q 1428 -91 1123 -41 \n", "Q 819 9 494 109 \n", "L 494 744 \n", "Q 775 591 1075 516 \n", "Q 1375 441 1709 441 \n", "Q 2250 441 2565 725 \n", "Q 2881 1009 2881 1497 \n", "Q 2881 1984 2565 2268 \n", "Q 2250 2553 1709 2553 \n", "Q 1456 2553 1204 2497 \n", "Q 953 2441 691 2322 \n", "L 691 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2d\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"36.083984\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"99.707031\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"131.494141\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_5\">\n", "     <g id=\"text_5\">\n", "      <!-- 0.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(193.898574 203.236563)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_6\">\n", "     <g id=\"text_6\">\n", "      <!-- 0.5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(215.68781 203.236563)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_7\">\n", "     <g id=\"text_7\">\n", "      <!-- 1.1 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(239.991186 203.236563)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-31\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_8\">\n", "     <g id=\"text_8\">\n", "      <!-- 1.9 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(275.608208 203.236563)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-31\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-39\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_9\">\n", "     <g id=\"text_9\">\n", "      <!-- 4.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(361.089063 203.236563)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-34\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"text_10\">\n", "     <!-- z -->\n", "     <g style=\"fill: #262626\" transform=\"translate(199.016406 216.914688)scale(0.1 -0.1)\">\n", "      <defs>\n", "       <path id=\"DejaVuSans-7a\" d=\"M 353 3500 \n", "L 3084 3500 \n", "L 3084 2975 \n", "L 922 459 \n", "L 3084 459 \n", "L 3084 0 \n", "L 275 0 \n", "L 275 525 \n", "L 2438 3041 \n", "L 353 3041 \n", "L 353 3500 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      </defs>\n", "      <use xlink:href=\"#DejaVuSans-7a\"/>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"matplotlib.axis_2\">\n", "    <g id=\"ytick_1\">\n", "     <g id=\"text_11\">\n", "      <!-- 0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(20.878125 192.437344)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_2\">\n", "     <g id=\"text_12\">\n", "      <!-- 1 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(20.878125 167.178983)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-31\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_3\">\n", "     <g id=\"text_13\">\n", "      <!-- 2 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(20.878125 141.920622)scale(0.1 -0.1)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-32\" d=\"M 1228 531 \n", "L 3431 531 \n", "L 3431 0 \n", "L 469 0 \n", "L 469 531 \n", "Q 828 903 1448 1529 \n", "Q 2069 2156 2228 2338 \n", "Q 2531 2678 2651 2914 \n", "Q 2772 3150 2772 3378 \n", "Q 2772 3750 2511 3984 \n", "Q 2250 4219 1831 4219 \n", "Q 1534 4219 1204 4116 \n", "Q 875 4013 500 3803 \n", "L 500 4441 \n", "Q 881 4594 1212 4672 \n", "Q 1544 4750 1819 4750 \n", "Q 2544 4750 2975 4387 \n", "Q 3406 4025 3406 3419 \n", "Q 3406 3131 3298 2873 \n", "Q 3191 2616 2906 2266 \n", "Q 2828 2175 2409 1742 \n", "Q 1991 1309 1228 531 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-32\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_4\">\n", "     <g id=\"text_14\">\n", "      <!-- 3 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(20.878125 116.662261)scale(0.1 -0.1)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-33\" d=\"M 2597 2516 \n", "Q 3050 2419 3304 2112 \n", "Q 3559 1806 3559 1356 \n", "Q 3559 666 3084 287 \n", "Q 2609 -91 1734 -91 \n", "Q 1441 -91 1130 -33 \n", "Q 819 25 488 141 \n", "L 488 750 \n", "Q 750 597 1062 519 \n", "Q 1375 441 1716 441 \n", "Q 2309 441 2620 675 \n", "Q 2931 909 2931 1356 \n", "Q 2931 1769 2642 2001 \n", "Q 2353 2234 1838 2234 \n", "L 1294 2234 \n", "L 1294 2753 \n", "L 1863 2753 \n", "Q 2328 2753 2575 2939 \n", "Q 2822 3125 2822 3475 \n", "Q 2822 3834 2567 4026 \n", "Q 2313 4219 1838 4219 \n", "Q 1578 4219 1281 4162 \n", "Q 984 4106 628 3988 \n", "L 628 4550 \n", "Q 988 4650 1302 4700 \n", "Q 1616 4750 1894 4750 \n", "Q 2613 4750 3031 4423 \n", "Q 3450 4097 3450 3541 \n", "Q 3450 3153 3228 2886 \n", "Q 3006 2619 2597 2516 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-33\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_5\">\n", "     <g id=\"text_15\">\n", "      <!-- 4 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(20.878125 91.4039)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-34\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_6\">\n", "     <g id=\"text_16\">\n", "      <!-- 5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(20.878125 66.145539)scale(0.1 -0.1)\">\n", "       <use xlink:href=\"#DejaVuSans-35\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_7\">\n", "     <g id=\"text_17\">\n", "      <!-- 6 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(20.878125 40.887178)scale(0.1 -0.1)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-36\" d=\"M 2113 2584 \n", "Q 1688 2584 1439 2293 \n", "Q 1191 2003 1191 1497 \n", "Q 1191 994 1439 701 \n", "Q 1688 409 2113 409 \n", "Q 2538 409 2786 701 \n", "Q 3034 994 3034 1497 \n", "Q 3034 2003 2786 2293 \n", "Q 2538 2584 2113 2584 \n", "z\n", "M 3366 4563 \n", "L 3366 3988 \n", "Q 3128 4100 2886 4159 \n", "Q 2644 4219 2406 4219 \n", "Q 1781 4219 1451 3797 \n", "Q 1122 3375 1075 2522 \n", "Q 1259 2794 1537 2939 \n", "Q 1816 3084 2150 3084 \n", "Q 2853 3084 3261 2657 \n", "Q 3669 2231 3669 1497 \n", "Q 3669 778 3244 343 \n", "Q 2819 -91 2113 -91 \n", "Q 1303 -91 875 529 \n", "Q 447 1150 447 2328 \n", "Q 447 3434 972 4092 \n", "Q 1497 4750 2381 4750 \n", "Q 2619 4750 2861 4703 \n", "Q 3103 4656 3366 4563 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-36\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"text_18\">\n", "     <!-- Probability -->\n", "     <g style=\"fill: #262626\" transform=\"translate(14.798437 131.907031)rotate(-90)scale(0.1 -0.1)\">\n", "      <defs>\n", "       <path id=\"DejaVuSans-50\" d=\"M 1259 4147 \n", "L 1259 2394 \n", "L 2053 2394 \n", "Q 2494 2394 2734 2622 \n", "Q 2975 2850 2975 3272 \n", "Q 2975 3691 2734 3919 \n", "Q 2494 4147 2053 4147 \n", "L 1259 4147 \n", "z\n", "M 628 4666 \n", "L 2053 4666 \n", "Q 2838 4666 3239 4311 \n", "Q 3641 3956 3641 3272 \n", "Q 3641 2581 3239 2228 \n", "Q 2838 1875 2053 1875 \n", "L 1259 1875 \n", "L 1259 0 \n", "L 628 0 \n", "L 628 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-72\" d=\"M 2631 2963 \n", "Q 2534 3019 2420 3045 \n", "Q 2306 3072 2169 3072 \n", "Q 1681 3072 1420 2755 \n", "Q 1159 2438 1159 1844 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 3500 \n", "L 1159 3500 \n", "L 1159 2956 \n", "Q 1341 3275 1631 3429 \n", "Q 1922 3584 2338 3584 \n", "Q 2397 3584 2469 3576 \n", "Q 2541 3569 2628 3553 \n", "L 2631 2963 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-6f\" d=\"M 1959 3097 \n", "Q 1497 3097 1228 2736 \n", "Q 959 2375 959 1747 \n", "Q 959 1119 1226 758 \n", "Q 1494 397 1959 397 \n", "Q 2419 397 2687 759 \n", "Q 2956 1122 2956 1747 \n", "Q 2956 2369 2687 2733 \n", "Q 2419 3097 1959 3097 \n", "z\n", "M 1959 3584 \n", "Q 2709 3584 3137 3096 \n", "Q 3566 2609 3566 1747 \n", "Q 3566 888 3137 398 \n", "Q 2709 -91 1959 -91 \n", "Q 1206 -91 779 398 \n", "Q 353 888 353 1747 \n", "Q 353 2609 779 3096 \n", "Q 1206 3584 1959 3584 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-62\" d=\"M 3116 1747 \n", "Q 3116 2381 2855 2742 \n", "Q 2594 3103 2138 3103 \n", "Q 1681 3103 1420 2742 \n", "Q 1159 2381 1159 1747 \n", "Q 1159 1113 1420 752 \n", "Q 1681 391 2138 391 \n", "Q 2594 391 2855 752 \n", "Q 3116 1113 3116 1747 \n", "z\n", "M 1159 2969 \n", "Q 1341 3281 1617 3432 \n", "Q 1894 3584 2278 3584 \n", "Q 2916 3584 3314 3078 \n", "Q 3713 2572 3713 1747 \n", "Q 3713 922 3314 415 \n", "Q 2916 -91 2278 -91 \n", "Q 1894 -91 1617 61 \n", "Q 1341 213 1159 525 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 4863 \n", "L 1159 4863 \n", "L 1159 2969 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-61\" d=\"M 2194 1759 \n", "Q 1497 1759 1228 1600 \n", "Q 959 1441 959 1056 \n", "Q 959 750 1161 570 \n", "Q 1363 391 1709 391 \n", "Q 2188 391 2477 730 \n", "Q 2766 1069 2766 1631 \n", "L 2766 1759 \n", "L 2194 1759 \n", "z\n", "M 3341 1997 \n", "L 3341 0 \n", "L 2766 0 \n", "L 2766 531 \n", "Q 2569 213 2275 61 \n", "Q 1981 -91 1556 -91 \n", "Q 1019 -91 701 211 \n", "Q 384 513 384 1019 \n", "Q 384 1609 779 1909 \n", "Q 1175 2209 1959 2209 \n", "L 2766 2209 \n", "L 2766 2266 \n", "Q 2766 2663 2505 2880 \n", "Q 2244 3097 1772 3097 \n", "Q 1472 3097 1187 3025 \n", "Q 903 2953 641 2809 \n", "L 641 3341 \n", "Q 956 3463 1253 3523 \n", "Q 1550 3584 1831 3584 \n", "Q 2591 3584 2966 3190 \n", "Q 3341 2797 3341 1997 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-69\" d=\"M 603 3500 \n", "L 1178 3500 \n", "L 1178 0 \n", "L 603 0 \n", "L 603 3500 \n", "z\n", "M 603 4863 \n", "L 1178 4863 \n", "L 1178 4134 \n", "L 603 4134 \n", "L 603 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-6c\" d=\"M 603 4863 \n", "L 1178 4863 \n", "L 1178 0 \n", "L 603 0 \n", "L 603 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-74\" d=\"M 1172 4494 \n", "L 1172 3500 \n", "L 2356 3500 \n", "L 2356 3053 \n", "L 1172 3053 \n", "L 1172 1153 \n", "Q 1172 725 1289 603 \n", "Q 1406 481 1766 481 \n", "L 2356 481 \n", "L 2356 0 \n", "L 1766 0 \n", "Q 1100 0 847 248 \n", "Q 594 497 594 1153 \n", "L 594 3053 \n", "L 172 3053 \n", "L 172 3500 \n", "L 594 3500 \n", "L 594 4494 \n", "L 1172 4494 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       <path id=\"DejaVuSans-79\" d=\"M 2059 -325 \n", "Q 1816 -950 1584 -1140 \n", "Q 1353 -1331 966 -1331 \n", "L 506 -1331 \n", "L 506 -850 \n", "L 844 -850 \n", "Q 1081 -850 1212 -737 \n", "Q 1344 -625 1503 -206 \n", "L 1606 56 \n", "L 191 3500 \n", "L 800 3500 \n", "L 1894 763 \n", "L 2988 3500 \n", "L 3597 3500 \n", "L 2059 -325 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      </defs>\n", "      <use xlink:href=\"#DejaVuSans-50\"/>\n", "      <use xlink:href=\"#DejaVuSans-72\" x=\"58.552734\"/>\n", "      <use xlink:href=\"#DejaVuSans-6f\" x=\"97.416016\"/>\n", "      <use xlink:href=\"#DejaVuSans-62\" x=\"158.597656\"/>\n", "      <use xlink:href=\"#DejaVuSans-61\" x=\"222.074219\"/>\n", "      <use xlink:href=\"#DejaVuSans-62\" x=\"283.353516\"/>\n", "      <use xlink:href=\"#DejaVuSans-69\" x=\"346.830078\"/>\n", "      <use xlink:href=\"#DejaVuSans-6c\" x=\"374.613281\"/>\n", "      <use xlink:href=\"#DejaVuSans-69\" x=\"402.396484\"/>\n", "      <use xlink:href=\"#DejaVuSans-74\" x=\"430.179688\"/>\n", "      <use xlink:href=\"#DejaVuSans-79\" x=\"469.388672\"/>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_1\">\n", "    <defs>\n", "     <path id=\"m86a3a2cffd\" d=\"M 34.240625 -37.55625 \n", "L 34.240625 -39.697679 \n", "L 34.659648 -39.718418 \n", "L 35.078672 -39.739352 \n", "L 35.497695 -39.76048 \n", "L 35.916719 -39.781804 \n", "L 36.335742 -39.803325 \n", "L 36.754765 -39.825048 \n", "L 37.173789 -39.846971 \n", "L 37.592812 -39.869097 \n", "L 38.011835 -39.891429 \n", "L 38.430859 -39.913966 \n", "L 38.849882 -39.936713 \n", "L 39.268906 -39.959668 \n", "L 39.687929 -39.982837 \n", "L 40.106952 -40.006218 \n", "L 40.525976 -40.029814 \n", "L 40.945009 -40.053628 \n", "L 41.364032 -40.077661 \n", "L 41.783056 -40.101914 \n", "L 42.202079 -40.12639 \n", "L 42.621103 -40.151089 \n", "L 43.040126 -40.176016 \n", "L 43.459149 -40.201171 \n", "L 43.878173 -40.226554 \n", "L 44.297196 -40.252171 \n", "L 44.716219 -40.27802 \n", "L 45.135243 -40.304104 \n", "L 45.554266 -40.330427 \n", "L 45.97329 -40.356988 \n", "L 46.392313 -40.383791 \n", "L 46.811336 -40.410837 \n", "L 47.23036 -40.438129 \n", "L 47.649383 -40.465667 \n", "L 48.068407 -40.493456 \n", "L 48.48743 -40.521493 \n", "L 48.906453 -40.549785 \n", "L 49.325477 -40.578331 \n", "L 49.7445 -40.607135 \n", "L 50.163523 -40.636197 \n", "L 50.582547 -40.665521 \n", "L 51.00158 -40.695109 \n", "L 51.420604 -40.72496 \n", "L 51.839627 -40.755079 \n", "L 52.25865 -40.785467 \n", "L 52.677674 -40.816126 \n", "L 53.096697 -40.847059 \n", "L 53.51572 -40.878268 \n", "L 53.934744 -40.909754 \n", "L 54.353767 -40.94152 \n", "L 54.772791 -40.973567 \n", "L 55.191814 -41.005899 \n", "L 55.610837 -41.038515 \n", "L 56.029861 -41.071422 \n", "L 56.448884 -41.104617 \n", "L 56.867907 -41.138106 \n", "L 57.286931 -41.17189 \n", "L 57.705954 -41.20597 \n", "L 58.124978 -41.24035 \n", "L 58.544001 -41.275029 \n", "L 58.963024 -41.310013 \n", "L 59.382048 -41.345302 \n", "L 59.801071 -41.3809 \n", "L 60.220095 -41.416808 \n", "L 60.639118 -41.453027 \n", "L 61.058151 -41.489563 \n", "L 61.477175 -41.526414 \n", "L 61.896198 -41.563585 \n", "L 62.315221 -41.601075 \n", "L 62.734245 -41.638891 \n", "L 63.153268 -41.67703 \n", "L 63.572292 -41.715499 \n", "L 63.991315 -41.754298 \n", "L 64.410338 -41.793431 \n", "L 64.829362 -41.832896 \n", "L 65.248385 -41.8727 \n", "L 65.667408 -41.912843 \n", "L 66.086432 -41.953329 \n", "L 66.505455 -41.994156 \n", "L 66.924479 -42.035331 \n", "L 67.343502 -42.076855 \n", "L 67.762525 -42.118732 \n", "L 68.181549 -42.160959 \n", "L 68.600572 -42.203541 \n", "L 69.019595 -42.246484 \n", "L 69.438619 -42.289785 \n", "L 69.857642 -42.33345 \n", "L 70.276666 -42.377478 \n", "L 70.695689 -42.421872 \n", "L 71.114722 -42.466639 \n", "L 71.533746 -42.511777 \n", "L 71.952769 -42.557286 \n", "L 72.371793 -42.603175 \n", "L 72.790816 -42.649442 \n", "L 73.209839 -42.696088 \n", "L 73.628863 -42.743117 \n", "L 74.047886 -42.790534 \n", "L 74.466909 -42.838337 \n", "L 74.885933 -42.88653 \n", "L 75.304956 -42.935117 \n", "L 75.72398 -42.984097 \n", "L 76.143003 -43.033475 \n", "L 76.562026 -43.083251 \n", "L 76.98105 -43.133429 \n", "L 77.400073 -43.184011 \n", "L 77.819096 -43.234998 \n", "L 78.23812 -43.286395 \n", "L 78.657143 -43.338203 \n", "L 79.076167 -43.390423 \n", "L 79.49519 -43.443058 \n", "L 79.914213 -43.496112 \n", "L 80.333237 -43.549584 \n", "L 80.75226 -43.603476 \n", "L 81.171283 -43.657792 \n", "L 81.590307 -43.712535 \n", "L 82.00933 -43.767706 \n", "L 82.428354 -43.823306 \n", "L 82.847377 -43.87934 \n", "L 83.2664 -43.935806 \n", "L 83.685424 -43.992711 \n", "L 84.104447 -44.050055 \n", "L 84.523481 -44.107837 \n", "L 84.942504 -44.166066 \n", "L 85.361527 -44.224735 \n", "L 85.780551 -44.283852 \n", "L 86.199574 -44.343419 \n", "L 86.618597 -44.403437 \n", "L 87.037621 -44.463907 \n", "L 87.456644 -44.524832 \n", "L 87.875668 -44.586214 \n", "L 88.294691 -44.648054 \n", "L 88.713714 -44.710354 \n", "L 89.132738 -44.773115 \n", "L 89.551761 -44.83634 \n", "L 89.970784 -44.900031 \n", "L 90.389808 -44.964191 \n", "L 90.808831 -45.028819 \n", "L 91.227855 -45.093917 \n", "L 91.646878 -45.159492 \n", "L 92.065901 -45.225535 \n", "L 92.484925 -45.292059 \n", "L 92.903948 -45.359056 \n", "L 93.322971 -45.426536 \n", "L 93.741995 -45.494493 \n", "L 94.161018 -45.562934 \n", "L 94.580052 -45.63186 \n", "L 94.999075 -45.701271 \n", "L 95.418098 -45.771167 \n", "L 95.837122 -45.841546 \n", "L 96.256145 -45.912418 \n", "L 96.675169 -45.983783 \n", "L 97.094192 -46.055632 \n", "L 97.513215 -46.127979 \n", "L 97.932239 -46.200819 \n", "L 98.351262 -46.274152 \n", "L 98.770285 -46.347981 \n", "L 99.189309 -46.422306 \n", "L 99.608332 -46.49713 \n", "L 100.027356 -46.572453 \n", "L 100.446379 -46.648271 \n", "L 100.865402 -46.724592 \n", "L 101.284426 -46.801416 \n", "L 101.703449 -46.878738 \n", "L 102.122472 -46.956567 \n", "L 102.541496 -47.034892 \n", "L 102.960519 -47.113725 \n", "L 103.379543 -47.193061 \n", "L 103.798566 -47.272898 \n", "L 104.217589 -47.353243 \n", "L 104.636623 -47.434093 \n", "L 105.055646 -47.515447 \n", "L 105.474669 -47.597304 \n", "L 105.893693 -47.67967 \n", "L 106.312716 -47.76254 \n", "L 106.73174 -47.845913 \n", "L 107.150763 -47.929794 \n", "L 107.569786 -48.014173 \n", "L 107.98881 -48.099064 \n", "L 108.407833 -48.184455 \n", "L 108.826857 -48.270355 \n", "L 109.24588 -48.356753 \n", "L 109.664903 -48.443656 \n", "L 110.083927 -48.531058 \n", "L 110.50295 -48.618963 \n", "L 110.921973 -48.707369 \n", "L 111.340997 -48.796278 \n", "L 111.76002 -48.885679 \n", "L 112.179044 -48.975585 \n", "L 112.598067 -49.065985 \n", "L 113.01709 -49.15688 \n", "L 113.436114 -49.248268 \n", "L 113.855137 -49.340148 \n", "L 114.27416 -49.432521 \n", "L 114.693194 -49.525383 \n", "L 115.112217 -49.618732 \n", "L 115.531241 -49.71257 \n", "L 115.950264 -49.806888 \n", "L 116.369287 -49.901691 \n", "L 116.788311 -49.996978 \n", "L 117.207334 -50.092738 \n", "L 117.626357 -50.188975 \n", "L 118.045381 -50.285684 \n", "L 118.464404 -50.382871 \n", "L 118.883428 -50.480516 \n", "L 119.302451 -50.578631 \n", "L 119.721474 -50.677216 \n", "L 120.140498 -50.776251 \n", "L 120.140498 -37.55625 \n", "L 120.140498 -37.55625 \n", "L 119.721474 -37.55625 \n", "L 119.302451 -37.55625 \n", "L 118.883428 -37.55625 \n", "L 118.464404 -37.55625 \n", "L 118.045381 -37.55625 \n", "L 117.626357 -37.55625 \n", "L 117.207334 -37.55625 \n", "L 116.788311 -37.55625 \n", "L 116.369287 -37.55625 \n", "L 115.950264 -37.55625 \n", "L 115.531241 -37.55625 \n", "L 115.112217 -37.55625 \n", "L 114.693194 -37.55625 \n", "L 114.27416 -37.55625 \n", "L 113.855137 -37.55625 \n", "L 113.436114 -37.55625 \n", "L 113.01709 -37.55625 \n", "L 112.598067 -37.55625 \n", "L 112.179044 -37.55625 \n", "L 111.76002 -37.55625 \n", "L 111.340997 -37.55625 \n", "L 110.921973 -37.55625 \n", "L 110.50295 -37.55625 \n", "L 110.083927 -37.55625 \n", "L 109.664903 -37.55625 \n", "L 109.24588 -37.55625 \n", "L 108.826857 -37.55625 \n", "L 108.407833 -37.55625 \n", "L 107.98881 -37.55625 \n", "L 107.569786 -37.55625 \n", "L 107.150763 -37.55625 \n", "L 106.73174 -37.55625 \n", "L 106.312716 -37.55625 \n", "L 105.893693 -37.55625 \n", "L 105.474669 -37.55625 \n", "L 105.055646 -37.55625 \n", "L 104.636623 -37.55625 \n", "L 104.217589 -37.55625 \n", "L 103.798566 -37.55625 \n", "L 103.379543 -37.55625 \n", "L 102.960519 -37.55625 \n", "L 102.541496 -37.55625 \n", "L 102.122472 -37.55625 \n", "L 101.703449 -37.55625 \n", "L 101.284426 -37.55625 \n", "L 100.865402 -37.55625 \n", "L 100.446379 -37.55625 \n", "L 100.027356 -37.55625 \n", "L 99.608332 -37.55625 \n", "L 99.189309 -37.55625 \n", "L 98.770285 -37.55625 \n", "L 98.351262 -37.55625 \n", "L 97.932239 -37.55625 \n", "L 97.513215 -37.55625 \n", "L 97.094192 -37.55625 \n", "L 96.675169 -37.55625 \n", "L 96.256145 -37.55625 \n", "L 95.837122 -37.55625 \n", "L 95.418098 -37.55625 \n", "L 94.999075 -37.55625 \n", "L 94.580052 -37.55625 \n", "L 94.161018 -37.55625 \n", "L 93.741995 -37.55625 \n", "L 93.322971 -37.55625 \n", "L 92.903948 -37.55625 \n", "L 92.484925 -37.55625 \n", "L 92.065901 -37.55625 \n", "L 91.646878 -37.55625 \n", "L 91.227855 -37.55625 \n", "L 90.808831 -37.55625 \n", "L 90.389808 -37.55625 \n", "L 89.970784 -37.55625 \n", "L 89.551761 -37.55625 \n", "L 89.132738 -37.55625 \n", "L 88.713714 -37.55625 \n", "L 88.294691 -37.55625 \n", "L 87.875668 -37.55625 \n", "L 87.456644 -37.55625 \n", "L 87.037621 -37.55625 \n", "L 86.618597 -37.55625 \n", "L 86.199574 -37.55625 \n", "L 85.780551 -37.55625 \n", "L 85.361527 -37.55625 \n", "L 84.942504 -37.55625 \n", "L 84.523481 -37.55625 \n", "L 84.104447 -37.55625 \n", "L 83.685424 -37.55625 \n", "L 83.2664 -37.55625 \n", "L 82.847377 -37.55625 \n", "L 82.428354 -37.55625 \n", "L 82.00933 -37.55625 \n", "L 81.590307 -37.55625 \n", "L 81.171283 -37.55625 \n", "L 80.75226 -37.55625 \n", "L 80.333237 -37.55625 \n", "L 79.914213 -37.55625 \n", "L 79.49519 -37.55625 \n", "L 79.076167 -37.55625 \n", "L 78.657143 -37.55625 \n", "L 78.23812 -37.55625 \n", "L 77.819096 -37.55625 \n", "L 77.400073 -37.55625 \n", "L 76.98105 -37.55625 \n", "L 76.562026 -37.55625 \n", "L 76.143003 -37.55625 \n", "L 75.72398 -37.55625 \n", "L 75.304956 -37.55625 \n", "L 74.885933 -37.55625 \n", "L 74.466909 -37.55625 \n", "L 74.047886 -37.55625 \n", "L 73.628863 -37.55625 \n", "L 73.209839 -37.55625 \n", "L 72.790816 -37.55625 \n", "L 72.371793 -37.55625 \n", "L 71.952769 -37.55625 \n", "L 71.533746 -37.55625 \n", "L 71.114722 -37.55625 \n", "L 70.695689 -37.55625 \n", "L 70.276666 -37.55625 \n", "L 69.857642 -37.55625 \n", "L 69.438619 -37.55625 \n", "L 69.019595 -37.55625 \n", "L 68.600572 -37.55625 \n", "L 68.181549 -37.55625 \n", "L 67.762525 -37.55625 \n", "L 67.343502 -37.55625 \n", "L 66.924479 -37.55625 \n", "L 66.505455 -37.55625 \n", "L 66.086432 -37.55625 \n", "L 65.667408 -37.55625 \n", "L 65.248385 -37.55625 \n", "L 64.829362 -37.55625 \n", "L 64.410338 -37.55625 \n", "L 63.991315 -37.55625 \n", "L 63.572292 -37.55625 \n", "L 63.153268 -37.55625 \n", "L 62.734245 -37.55625 \n", "L 62.315221 -37.55625 \n", "L 61.896198 -37.55625 \n", "L 61.477175 -37.55625 \n", "L 61.058151 -37.55625 \n", "L 60.639118 -37.55625 \n", "L 60.220095 -37.55625 \n", "L 59.801071 -37.55625 \n", "L 59.382048 -37.55625 \n", "L 58.963024 -37.55625 \n", "L 58.544001 -37.55625 \n", "L 58.124978 -37.55625 \n", "L 57.705954 -37.55625 \n", "L 57.286931 -37.55625 \n", "L 56.867907 -37.55625 \n", "L 56.448884 -37.55625 \n", "L 56.029861 -37.55625 \n", "L 55.610837 -37.55625 \n", "L 55.191814 -37.55625 \n", "L 54.772791 -37.55625 \n", "L 54.353767 -37.55625 \n", "L 53.934744 -37.55625 \n", "L 53.51572 -37.55625 \n", "L 53.096697 -37.55625 \n", "L 52.677674 -37.55625 \n", "L 52.25865 -37.55625 \n", "L 51.839627 -37.55625 \n", "L 51.420604 -37.55625 \n", "L 51.00158 -37.55625 \n", "L 50.582547 -37.55625 \n", "L 50.163523 -37.55625 \n", "L 49.7445 -37.55625 \n", "L 49.325477 -37.55625 \n", "L 48.906453 -37.55625 \n", "L 48.48743 -37.55625 \n", "L 48.068407 -37.55625 \n", "L 47.649383 -37.55625 \n", "L 47.23036 -37.55625 \n", "L 46.811336 -37.55625 \n", "L 46.392313 -37.55625 \n", "L 45.97329 -37.55625 \n", "L 45.554266 -37.55625 \n", "L 45.135243 -37.55625 \n", "L 44.716219 -37.55625 \n", "L 44.297196 -37.55625 \n", "L 43.878173 -37.55625 \n", "L 43.459149 -37.55625 \n", "L 43.040126 -37.55625 \n", "L 42.621103 -37.55625 \n", "L 42.202079 -37.55625 \n", "L 41.783056 -37.55625 \n", "L 41.364032 -37.55625 \n", "L 40.945009 -37.55625 \n", "L 40.525976 -37.55625 \n", "L 40.106952 -37.55625 \n", "L 39.687929 -37.55625 \n", "L 39.268906 -37.55625 \n", "L 38.849882 -37.55625 \n", "L 38.430859 -37.55625 \n", "L 38.011835 -37.55625 \n", "L 37.592812 -37.55625 \n", "L 37.173789 -37.55625 \n", "L 36.754765 -37.55625 \n", "L 36.335742 -37.55625 \n", "L 35.916719 -37.55625 \n", "L 35.497695 -37.55625 \n", "L 35.078672 -37.55625 \n", "L 34.659648 -37.55625 \n", "L 34.240625 -37.55625 \n", "z\n", "\" style=\"stroke: #1f77b4; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pfb33315d50)\">\n", "     <use xlink:href=\"#m86a3a2cffd\" x=\"0\" y=\"226.194375\" style=\"fill: #1f77b4; fill-opacity: 0.5; stroke: #1f77b4; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_2\">\n", "    <defs>\n", "     <path id=\"m262a478bf3\" d=\"M 120.559521 -37.55625 \n", "L 120.559521 -73.07491 \n", "L 120.978549 -73.341453 \n", "L 121.397573 -73.609189 \n", "L 121.816596 -73.878115 \n", "L 122.23562 -74.148227 \n", "L 122.654643 -74.419504 \n", "L 123.073666 -74.691959 \n", "L 123.49269 -74.965553 \n", "L 123.911713 -75.24032 \n", "L 124.330742 -75.516198 \n", "L 124.74976 -75.793204 \n", "L 125.168783 -76.071312 \n", "L 125.587807 -76.350519 \n", "L 126.00683 -76.63082 \n", "L 126.425853 -76.912204 \n", "L 126.844877 -77.194634 \n", "L 127.2639 -77.478119 \n", "L 127.682929 -77.762635 \n", "L 128.101952 -78.048175 \n", "L 128.520975 -78.334719 \n", "L 128.939999 -78.622246 \n", "L 129.359022 -78.910765 \n", "L 129.778045 -79.200239 \n", "L 130.197069 -79.490656 \n", "L 130.616092 -79.782003 \n", "L 131.035121 -80.074261 \n", "L 131.454144 -80.367439 \n", "L 131.873167 -80.661465 \n", "L 132.292191 -80.956373 \n", "L 132.711214 -81.252114 \n", "L 133.130237 -81.548689 \n", "L 133.549261 -81.846073 \n", "L 133.968284 -82.144242 \n", "L 134.387313 -82.443187 \n", "L 134.806331 -82.742869 \n", "L 135.225354 -83.043283 \n", "L 135.644378 -83.3444 \n", "L 136.063401 -83.646221 \n", "L 136.482425 -83.948697 \n", "L 136.901448 -84.251819 \n", "L 137.320471 -84.555567 \n", "L 137.7395 -84.859932 \n", "L 138.158523 -85.164855 \n", "L 138.577546 -85.470338 \n", "L 138.99657 -85.776345 \n", "L 139.415593 -86.082859 \n", "L 139.834617 -86.389868 \n", "L 140.25364 -86.697331 \n", "L 140.672663 -87.005217 \n", "L 141.091692 -87.313522 \n", "L 141.51071 -87.622199 \n", "L 141.929733 -87.931222 \n", "L 142.348757 -88.240593 \n", "L 142.76778 -88.550247 \n", "L 143.186804 -88.86019 \n", "L 143.605827 -89.17037 \n", "L 144.02485 -89.480762 \n", "L 144.443879 -89.79136 \n", "L 144.862902 -90.102108 \n", "L 145.281925 -90.412991 \n", "L 145.700949 -90.723975 \n", "L 146.119972 -91.035017 \n", "L 146.538996 -91.346108 \n", "L 146.958019 -91.657217 \n", "L 147.377042 -91.968293 \n", "L 147.796071 -92.279315 \n", "L 148.215094 -92.590256 \n", "L 148.634118 -92.901076 \n", "L 149.053141 -93.211757 \n", "L 149.472164 -93.522235 \n", "L 149.891188 -93.832511 \n", "L 150.310211 -94.142537 \n", "L 150.729234 -94.452273 \n", "L 151.148263 -94.761696 \n", "L 151.567281 -95.070773 \n", "L 151.986305 -95.379459 \n", "L 152.405328 -95.687711 \n", "L 152.824351 -95.995511 \n", "L 153.243375 -96.302814 \n", "L 153.662398 -96.609597 \n", "L 154.081421 -96.915812 \n", "L 154.50045 -97.221415 \n", "L 154.919473 -97.526387 \n", "L 155.338497 -97.830674 \n", "L 155.75752 -98.134254 \n", "L 155.75752 -37.55625 \n", "L 155.75752 -37.55625 \n", "L 155.338497 -37.55625 \n", "L 154.919473 -37.55625 \n", "L 154.50045 -37.55625 \n", "L 154.081421 -37.55625 \n", "L 153.662398 -37.55625 \n", "L 153.243375 -37.55625 \n", "L 152.824351 -37.55625 \n", "L 152.405328 -37.55625 \n", "L 151.986305 -37.55625 \n", "L 151.567281 -37.55625 \n", "L 151.148263 -37.55625 \n", "L 150.729234 -37.55625 \n", "L 150.310211 -37.55625 \n", "L 149.891188 -37.55625 \n", "L 149.472164 -37.55625 \n", "L 149.053141 -37.55625 \n", "L 148.634118 -37.55625 \n", "L 148.215094 -37.55625 \n", "L 147.796071 -37.55625 \n", "L 147.377042 -37.55625 \n", "L 146.958019 -37.55625 \n", "L 146.538996 -37.55625 \n", "L 146.119972 -37.55625 \n", "L 145.700949 -37.55625 \n", "L 145.281925 -37.55625 \n", "L 144.862902 -37.55625 \n", "L 144.443879 -37.55625 \n", "L 144.02485 -37.55625 \n", "L 143.605827 -37.55625 \n", "L 143.186804 -37.55625 \n", "L 142.76778 -37.55625 \n", "L 142.348757 -37.55625 \n", "L 141.929733 -37.55625 \n", "L 141.51071 -37.55625 \n", "L 141.091692 -37.55625 \n", "L 140.672663 -37.55625 \n", "L 140.25364 -37.55625 \n", "L 139.834617 -37.55625 \n", "L 139.415593 -37.55625 \n", "L 138.99657 -37.55625 \n", "L 138.577546 -37.55625 \n", "L 138.158523 -37.55625 \n", "L 137.7395 -37.55625 \n", "L 137.320471 -37.55625 \n", "L 136.901448 -37.55625 \n", "L 136.482425 -37.55625 \n", "L 136.063401 -37.55625 \n", "L 135.644378 -37.55625 \n", "L 135.225354 -37.55625 \n", "L 134.806331 -37.55625 \n", "L 134.387313 -37.55625 \n", "L 133.968284 -37.55625 \n", "L 133.549261 -37.55625 \n", "L 133.130237 -37.55625 \n", "L 132.711214 -37.55625 \n", "L 132.292191 -37.55625 \n", "L 131.873167 -37.55625 \n", "L 131.454144 -37.55625 \n", "L 131.035121 -37.55625 \n", "L 130.616092 -37.55625 \n", "L 130.197069 -37.55625 \n", "L 129.778045 -37.55625 \n", "L 129.359022 -37.55625 \n", "L 128.939999 -37.55625 \n", "L 128.520975 -37.55625 \n", "L 128.101952 -37.55625 \n", "L 127.682929 -37.55625 \n", "L 127.2639 -37.55625 \n", "L 126.844877 -37.55625 \n", "L 126.425853 -37.55625 \n", "L 126.00683 -37.55625 \n", "L 125.587807 -37.55625 \n", "L 125.168783 -37.55625 \n", "L 124.74976 -37.55625 \n", "L 124.330742 -37.55625 \n", "L 123.911713 -37.55625 \n", "L 123.49269 -37.55625 \n", "L 123.073666 -37.55625 \n", "L 122.654643 -37.55625 \n", "L 122.23562 -37.55625 \n", "L 121.816596 -37.55625 \n", "L 121.397573 -37.55625 \n", "L 120.978549 -37.55625 \n", "L 120.559521 -37.55625 \n", "z\n", "\" style=\"stroke: #ff7f0e; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pfb33315d50)\">\n", "     <use xlink:href=\"#m262a478bf3\" x=\"0\" y=\"226.194375\" style=\"fill: #ff7f0e; fill-opacity: 0.5; stroke: #ff7f0e; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_3\">\n", "    <defs>\n", "     <path id=\"mf547a6d129\" d=\"M 156.176543 -37.55625 \n", "L 156.176543 -159.317884 \n", "L 156.595567 -159.921959 \n", "L 157.01459 -160.524368 \n", "L 157.433613 -161.125023 \n", "L 157.852642 -161.723895 \n", "L 158.271665 -162.32087 \n", "L 158.690689 -162.915869 \n", "L 159.109712 -163.508815 \n", "L 159.528735 -164.099671 \n", "L 159.947761 -164.688301 \n", "L 160.366785 -165.274648 \n", "L 160.785808 -165.858615 \n", "L 161.204831 -166.440153 \n", "L 161.623855 -167.019177 \n", "L 162.042878 -167.595551 \n", "L 162.461902 -168.169275 \n", "L 162.880925 -168.740234 \n", "L 163.299951 -169.308331 \n", "L 163.718974 -169.87348 \n", "L 164.137998 -170.435642 \n", "L 164.557021 -170.994682 \n", "L 164.976044 -171.550581 \n", "L 165.395068 -172.103185 \n", "L 165.814091 -172.652455 \n", "L 166.233114 -173.198314 \n", "L 166.65214 -173.740647 \n", "L 167.071164 -174.279366 \n", "L 167.490187 -174.814453 \n", "L 167.90921 -175.345782 \n", "L 168.328234 -175.873276 \n", "L 168.747257 -176.396819 \n", "L 169.166281 -176.916364 \n", "L 169.585304 -177.431824 \n", "L 170.00433 -177.94316 \n", "L 170.423353 -178.450217 \n", "L 170.842377 -178.952939 \n", "L 171.2614 -179.451267 \n", "L 171.680423 -179.945067 \n", "L 172.099447 -180.434318 \n", "L 172.51847 -180.918926 \n", "L 172.937494 -181.398764 \n", "L 173.356519 -181.873813 \n", "L 173.775543 -182.343967 \n", "L 174.194566 -182.80915 \n", "L 174.61359 -183.269255 \n", "L 175.032615 -183.724253 \n", "L 175.451639 -184.17401 \n", "L 175.870662 -184.618525 \n", "L 176.289686 -185.057663 \n", "L 176.708711 -185.491329 \n", "L 177.127735 -185.919531 \n", "L 177.546758 -186.342126 \n", "L 177.965782 -186.759064 \n", "L 178.384805 -187.170241 \n", "L 178.803828 -187.575656 \n", "L 179.222852 -187.975164 \n", "L 179.641875 -188.368699 \n", "L 180.060901 -188.75625 \n", "L 180.060901 -37.55625 \n", "L 180.060901 -37.55625 \n", "L 179.641875 -37.55625 \n", "L 179.222852 -37.55625 \n", "L 178.803828 -37.55625 \n", "L 178.384805 -37.55625 \n", "L 177.965782 -37.55625 \n", "L 177.546758 -37.55625 \n", "L 177.127735 -37.55625 \n", "L 176.708711 -37.55625 \n", "L 176.289686 -37.55625 \n", "L 175.870662 -37.55625 \n", "L 175.451639 -37.55625 \n", "L 175.032615 -37.55625 \n", "L 174.61359 -37.55625 \n", "L 174.194566 -37.55625 \n", "L 173.775543 -37.55625 \n", "L 173.356519 -37.55625 \n", "L 172.937494 -37.55625 \n", "L 172.51847 -37.55625 \n", "L 172.099447 -37.55625 \n", "L 171.680423 -37.55625 \n", "L 171.2614 -37.55625 \n", "L 170.842377 -37.55625 \n", "L 170.423353 -37.55625 \n", "L 170.00433 -37.55625 \n", "L 169.585304 -37.55625 \n", "L 169.166281 -37.55625 \n", "L 168.747257 -37.55625 \n", "L 168.328234 -37.55625 \n", "L 167.90921 -37.55625 \n", "L 167.490187 -37.55625 \n", "L 167.071164 -37.55625 \n", "L 166.65214 -37.55625 \n", "L 166.233114 -37.55625 \n", "L 165.814091 -37.55625 \n", "L 165.395068 -37.55625 \n", "L 164.976044 -37.55625 \n", "L 164.557021 -37.55625 \n", "L 164.137998 -37.55625 \n", "L 163.718974 -37.55625 \n", "L 163.299951 -37.55625 \n", "L 162.880925 -37.55625 \n", "L 162.461902 -37.55625 \n", "L 162.042878 -37.55625 \n", "L 161.623855 -37.55625 \n", "L 161.204831 -37.55625 \n", "L 160.785808 -37.55625 \n", "L 160.366785 -37.55625 \n", "L 159.947761 -37.55625 \n", "L 159.528735 -37.55625 \n", "L 159.109712 -37.55625 \n", "L 158.690689 -37.55625 \n", "L 158.271665 -37.55625 \n", "L 157.852642 -37.55625 \n", "L 157.433613 -37.55625 \n", "L 157.01459 -37.55625 \n", "L 156.595567 -37.55625 \n", "L 156.176543 -37.55625 \n", "z\n", "\" style=\"stroke: #2ca02c; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pfb33315d50)\">\n", "     <use xlink:href=\"#mf547a6d129\" x=\"0\" y=\"226.194375\" style=\"fill: #2ca02c; fill-opacity: 0.5; stroke: #2ca02c; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_4\">\n", "    <defs>\n", "     <path id=\"m079bb74f88\" d=\"M 180.479924 -37.55625 \n", "L 180.479924 -113.346966 \n", "L 180.898948 -113.534599 \n", "L 181.317971 -113.71912 \n", "L 181.736996 -113.9005 \n", "L 182.156019 -114.07869 \n", "L 182.575044 -114.253653 \n", "L 182.994067 -114.425388 \n", "L 183.413091 -114.593851 \n", "L 183.832115 -114.758986 \n", "L 184.251139 -114.920772 \n", "L 184.670162 -115.079205 \n", "L 185.089185 -115.234237 \n", "L 185.508209 -115.38583 \n", "L 185.927233 -115.533972 \n", "L 186.346257 -115.678627 \n", "L 186.76528 -115.81976 \n", "L 187.184305 -115.957357 \n", "L 187.603328 -116.091398 \n", "L 188.022351 -116.221836 \n", "L 188.441376 -116.348651 \n", "L 188.860399 -116.471829 \n", "L 189.279424 -116.59134 \n", "L 189.698447 -116.707162 \n", "L 190.117471 -116.819278 \n", "L 190.536495 -116.927656 \n", "L 190.955519 -117.032276 \n", "L 191.374542 -117.133129 \n", "L 191.793566 -117.230185 \n", "L 192.21259 -117.323431 \n", "L 192.631614 -117.412837 \n", "L 193.050638 -117.498398 \n", "L 193.469662 -117.580101 \n", "L 193.888685 -117.657906 \n", "L 194.307709 -117.731828 \n", "L 194.726732 -117.801833 \n", "L 195.145756 -117.867912 \n", "L 195.56478 -117.93005 \n", "L 195.983804 -117.988242 \n", "L 196.402828 -118.04247 \n", "L 196.821851 -118.092723 \n", "L 197.240875 -118.139006 \n", "L 197.659899 -118.181271 \n", "L 198.078923 -118.219552 \n", "L 198.497947 -118.25382 \n", "L 198.91697 -118.284075 \n", "L 199.335994 -118.310292 \n", "L 199.755018 -118.332502 \n", "L 200.174042 -118.350674 \n", "L 200.593065 -118.364823 \n", "L 201.012089 -118.374921 \n", "L 201.431113 -118.380982 \n", "L 201.431113 -37.55625 \n", "L 201.431113 -37.55625 \n", "L 201.012089 -37.55625 \n", "L 200.593065 -37.55625 \n", "L 200.174042 -37.55625 \n", "L 199.755018 -37.55625 \n", "L 199.335994 -37.55625 \n", "L 198.91697 -37.55625 \n", "L 198.497947 -37.55625 \n", "L 198.078923 -37.55625 \n", "L 197.659899 -37.55625 \n", "L 197.240875 -37.55625 \n", "L 196.821851 -37.55625 \n", "L 196.402828 -37.55625 \n", "L 195.983804 -37.55625 \n", "L 195.56478 -37.55625 \n", "L 195.145756 -37.55625 \n", "L 194.726732 -37.55625 \n", "L 194.307709 -37.55625 \n", "L 193.888685 -37.55625 \n", "L 193.469662 -37.55625 \n", "L 193.050638 -37.55625 \n", "L 192.631614 -37.55625 \n", "L 192.21259 -37.55625 \n", "L 191.793566 -37.55625 \n", "L 191.374542 -37.55625 \n", "L 190.955519 -37.55625 \n", "L 190.536495 -37.55625 \n", "L 190.117471 -37.55625 \n", "L 189.698447 -37.55625 \n", "L 189.279424 -37.55625 \n", "L 188.860399 -37.55625 \n", "L 188.441376 -37.55625 \n", "L 188.022351 -37.55625 \n", "L 187.603328 -37.55625 \n", "L 187.184305 -37.55625 \n", "L 186.76528 -37.55625 \n", "L 186.346257 -37.55625 \n", "L 185.927233 -37.55625 \n", "L 185.508209 -37.55625 \n", "L 185.089185 -37.55625 \n", "L 184.670162 -37.55625 \n", "L 184.251139 -37.55625 \n", "L 183.832115 -37.55625 \n", "L 183.413091 -37.55625 \n", "L 182.994067 -37.55625 \n", "L 182.575044 -37.55625 \n", "L 182.156019 -37.55625 \n", "L 181.736996 -37.55625 \n", "L 181.317971 -37.55625 \n", "L 180.898948 -37.55625 \n", "L 180.479924 -37.55625 \n", "z\n", "\" style=\"stroke: #d62728; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pfb33315d50)\">\n", "     <use xlink:href=\"#m079bb74f88\" x=\"0\" y=\"226.194375\" style=\"fill: #d62728; fill-opacity: 0.5; stroke: #d62728; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_5\">\n", "    <defs>\n", "     <path id=\"m4d8781ee5e\" d=\"M 201.850137 -37.55625 \n", "L 201.850137 -67.866283 \n", "L 202.26916 -67.865524 \n", "L 202.688184 -67.863252 \n", "L 203.107208 -67.859465 \n", "L 203.526232 -67.854163 \n", "L 203.945256 -67.847348 \n", "L 204.364279 -67.839023 \n", "L 204.783303 -67.829184 \n", "L 205.202327 -67.817839 \n", "L 205.621351 -67.804988 \n", "L 206.040375 -67.790633 \n", "L 206.459398 -67.77478 \n", "L 206.878422 -67.757427 \n", "L 207.297446 -67.738582 \n", "L 207.71647 -67.718247 \n", "L 208.135493 -67.696425 \n", "L 208.554517 -67.673127 \n", "L 208.973541 -67.648344 \n", "L 209.392564 -67.622092 \n", "L 209.811588 -67.594371 \n", "L 210.230612 -67.56519 \n", "L 210.649636 -67.534555 \n", "L 211.06866 -67.50247 \n", "L 211.487683 -67.468939 \n", "L 211.906707 -67.433976 \n", "L 212.325731 -67.397579 \n", "L 212.744755 -67.35976 \n", "L 213.163778 -67.320527 \n", "L 213.582803 -67.279885 \n", "L 214.001826 -67.237842 \n", "L 214.420849 -67.194409 \n", "L 214.839874 -67.149592 \n", "L 215.258897 -67.103401 \n", "L 215.677921 -67.055845 \n", "L 216.096945 -67.006927 \n", "L 216.515969 -66.956669 \n", "L 216.934992 -66.905066 \n", "L 217.354017 -66.852141 \n", "L 217.77304 -66.797896 \n", "L 218.192063 -66.742342 \n", "L 218.611088 -66.685495 \n", "L 219.030111 -66.62736 \n", "L 219.449136 -66.567949 \n", "L 219.868159 -66.507276 \n", "L 220.287183 -66.445347 \n", "L 220.706208 -66.382177 \n", "L 221.125231 -66.31778 \n", "L 221.544254 -66.252163 \n", "L 221.963278 -66.185344 \n", "L 222.382301 -66.117326 \n", "L 222.801326 -66.048131 \n", "L 223.220349 -65.977769 \n", "L 223.220349 -37.55625 \n", "L 223.220349 -37.55625 \n", "L 222.801326 -37.55625 \n", "L 222.382301 -37.55625 \n", "L 221.963278 -37.55625 \n", "L 221.544254 -37.55625 \n", "L 221.125231 -37.55625 \n", "L 220.706208 -37.55625 \n", "L 220.287183 -37.55625 \n", "L 219.868159 -37.55625 \n", "L 219.449136 -37.55625 \n", "L 219.030111 -37.55625 \n", "L 218.611088 -37.55625 \n", "L 218.192063 -37.55625 \n", "L 217.77304 -37.55625 \n", "L 217.354017 -37.55625 \n", "L 216.934992 -37.55625 \n", "L 216.515969 -37.55625 \n", "L 216.096945 -37.55625 \n", "L 215.677921 -37.55625 \n", "L 215.258897 -37.55625 \n", "L 214.839874 -37.55625 \n", "L 214.420849 -37.55625 \n", "L 214.001826 -37.55625 \n", "L 213.582803 -37.55625 \n", "L 213.163778 -37.55625 \n", "L 212.744755 -37.55625 \n", "L 212.325731 -37.55625 \n", "L 211.906707 -37.55625 \n", "L 211.487683 -37.55625 \n", "L 211.06866 -37.55625 \n", "L 210.649636 -37.55625 \n", "L 210.230612 -37.55625 \n", "L 209.811588 -37.55625 \n", "L 209.392564 -37.55625 \n", "L 208.973541 -37.55625 \n", "L 208.554517 -37.55625 \n", "L 208.135493 -37.55625 \n", "L 207.71647 -37.55625 \n", "L 207.297446 -37.55625 \n", "L 206.878422 -37.55625 \n", "L 206.459398 -37.55625 \n", "L 206.040375 -37.55625 \n", "L 205.621351 -37.55625 \n", "L 205.202327 -37.55625 \n", "L 204.783303 -37.55625 \n", "L 204.364279 -37.55625 \n", "L 203.945256 -37.55625 \n", "L 203.526232 -37.55625 \n", "L 203.107208 -37.55625 \n", "L 202.688184 -37.55625 \n", "L 202.26916 -37.55625 \n", "L 201.850137 -37.55625 \n", "z\n", "\" style=\"stroke: #9467bd; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pfb33315d50)\">\n", "     <use xlink:href=\"#m4d8781ee5e\" x=\"0\" y=\"226.194375\" style=\"fill: #9467bd; fill-opacity: 0.5; stroke: #9467bd; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_6\">\n", "    <defs>\n", "     <path id=\"me6b72ba6eb\" d=\"M 223.639372 -37.55625 \n", "L 223.639372 -47.006249 \n", "L 224.058396 -46.982028 \n", "L 224.477419 -46.957432 \n", "L 224.896445 -46.932463 \n", "L 225.315468 -46.907126 \n", "L 225.734492 -46.881426 \n", "L 226.153515 -46.855367 \n", "L 226.572539 -46.828955 \n", "L 226.991564 -46.802195 \n", "L 227.410588 -46.775087 \n", "L 227.829611 -46.747642 \n", "L 228.248635 -46.719861 \n", "L 228.667658 -46.691749 \n", "L 229.086681 -46.663313 \n", "L 229.505705 -46.634555 \n", "L 229.924728 -46.605481 \n", "L 230.343754 -46.576098 \n", "L 230.762777 -46.546407 \n", "L 231.181801 -46.516416 \n", "L 231.600824 -46.486129 \n", "L 232.01985 -46.455551 \n", "L 232.438873 -46.424687 \n", "L 232.857897 -46.393543 \n", "L 233.27692 -46.362123 \n", "L 233.695946 -46.330432 \n", "L 234.114969 -46.298475 \n", "L 234.533993 -46.266257 \n", "L 234.953016 -46.233786 \n", "L 235.37204 -46.201064 \n", "L 235.791063 -46.168096 \n", "L 236.210086 -46.134889 \n", "L 236.62911 -46.101447 \n", "L 237.048136 -46.067775 \n", "L 237.467159 -46.033878 \n", "L 237.886182 -45.999762 \n", "L 238.305206 -45.965433 \n", "L 238.724229 -45.930895 \n", "L 239.143252 -45.896153 \n", "L 239.562276 -45.861211 \n", "L 239.981299 -45.826077 \n", "L 240.400325 -45.790754 \n", "L 240.819348 -45.755248 \n", "L 241.238372 -45.719564 \n", "L 241.657395 -45.683706 \n", "L 242.076419 -45.647681 \n", "L 242.495442 -45.611494 \n", "L 242.914465 -45.575148 \n", "L 243.333489 -45.538649 \n", "L 243.752515 -45.502002 \n", "L 244.171538 -45.465213 \n", "L 244.590561 -45.428286 \n", "L 245.009585 -45.391226 \n", "L 245.428608 -45.354039 \n", "L 245.847632 -45.316728 \n", "L 246.266655 -45.2793 \n", "L 246.685678 -45.241757 \n", "L 247.104702 -45.204107 \n", "L 247.523725 -45.166353 \n", "L 247.523725 -37.55625 \n", "L 247.523725 -37.55625 \n", "L 247.104702 -37.55625 \n", "L 246.685678 -37.55625 \n", "L 246.266655 -37.55625 \n", "L 245.847632 -37.55625 \n", "L 245.428608 -37.55625 \n", "L 245.009585 -37.55625 \n", "L 244.590561 -37.55625 \n", "L 244.171538 -37.55625 \n", "L 243.752515 -37.55625 \n", "L 243.333489 -37.55625 \n", "L 242.914465 -37.55625 \n", "L 242.495442 -37.55625 \n", "L 242.076419 -37.55625 \n", "L 241.657395 -37.55625 \n", "L 241.238372 -37.55625 \n", "L 240.819348 -37.55625 \n", "L 240.400325 -37.55625 \n", "L 239.981299 -37.55625 \n", "L 239.562276 -37.55625 \n", "L 239.143252 -37.55625 \n", "L 238.724229 -37.55625 \n", "L 238.305206 -37.55625 \n", "L 237.886182 -37.55625 \n", "L 237.467159 -37.55625 \n", "L 237.048136 -37.55625 \n", "L 236.62911 -37.55625 \n", "L 236.210086 -37.55625 \n", "L 235.791063 -37.55625 \n", "L 235.37204 -37.55625 \n", "L 234.953016 -37.55625 \n", "L 234.533993 -37.55625 \n", "L 234.114969 -37.55625 \n", "L 233.695946 -37.55625 \n", "L 233.27692 -37.55625 \n", "L 232.857897 -37.55625 \n", "L 232.438873 -37.55625 \n", "L 232.01985 -37.55625 \n", "L 231.600824 -37.55625 \n", "L 231.181801 -37.55625 \n", "L 230.762777 -37.55625 \n", "L 230.343754 -37.55625 \n", "L 229.924728 -37.55625 \n", "L 229.505705 -37.55625 \n", "L 229.086681 -37.55625 \n", "L 228.667658 -37.55625 \n", "L 228.248635 -37.55625 \n", "L 227.829611 -37.55625 \n", "L 227.410588 -37.55625 \n", "L 226.991564 -37.55625 \n", "L 226.572539 -37.55625 \n", "L 226.153515 -37.55625 \n", "L 225.734492 -37.55625 \n", "L 225.315468 -37.55625 \n", "L 224.896445 -37.55625 \n", "L 224.477419 -37.55625 \n", "L 224.058396 -37.55625 \n", "L 223.639372 -37.55625 \n", "z\n", "\" style=\"stroke: #8c564b; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pfb33315d50)\">\n", "     <use xlink:href=\"#me6b72ba6eb\" x=\"0\" y=\"226.194375\" style=\"fill: #8c564b; fill-opacity: 0.5; stroke: #8c564b; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_7\">\n", "    <defs>\n", "     <path id=\"m34741f690e\" d=\"M 247.942748 -37.55625 \n", "L 247.942748 -41.342375 \n", "L 248.361777 -41.323401 \n", "L 248.7808 -41.304383 \n", "L 249.199824 -41.285323 \n", "L 249.618847 -41.266223 \n", "L 250.03787 -41.247084 \n", "L 250.456894 -41.227911 \n", "L 250.875917 -41.208704 \n", "L 251.29494 -41.189467 \n", "L 251.713969 -41.1702 \n", "L 252.132992 -41.150908 \n", "L 252.552016 -41.131591 \n", "L 252.971039 -41.112252 \n", "L 253.390062 -41.092893 \n", "L 253.809086 -41.073517 \n", "L 254.228109 -41.054125 \n", "L 254.647132 -41.034719 \n", "L 255.066161 -41.015302 \n", "L 255.485179 -40.995875 \n", "L 255.904203 -40.976442 \n", "L 256.323226 -40.957003 \n", "L 256.742249 -40.93756 \n", "L 257.161273 -40.918116 \n", "L 257.580296 -40.898673 \n", "L 257.99932 -40.879233 \n", "L 258.418348 -40.859796 \n", "L 258.837371 -40.840366 \n", "L 259.256395 -40.820944 \n", "L 259.675418 -40.801533 \n", "L 260.094441 -40.782133 \n", "L 260.513465 -40.762746 \n", "L 260.932488 -40.743375 \n", "L 261.351512 -40.724021 \n", "L 261.77054 -40.704686 \n", "L 262.189563 -40.685372 \n", "L 262.608587 -40.666079 \n", "L 263.02761 -40.64681 \n", "L 263.446633 -40.627567 \n", "L 263.865657 -40.608351 \n", "L 264.28468 -40.589163 \n", "L 264.703704 -40.570006 \n", "L 265.122732 -40.55088 \n", "L 265.54175 -40.531788 \n", "L 265.960774 -40.51273 \n", "L 266.379797 -40.493708 \n", "L 266.79882 -40.474724 \n", "L 267.217844 -40.455778 \n", "L 267.636867 -40.436873 \n", "L 268.055891 -40.41801 \n", "L 268.474919 -40.399189 \n", "L 268.893942 -40.380413 \n", "L 269.312966 -40.361683 \n", "L 269.731989 -40.342999 \n", "L 270.151013 -40.324364 \n", "L 270.570036 -40.305778 \n", "L 270.989059 -40.287242 \n", "L 271.408083 -40.268758 \n", "L 271.827111 -40.250326 \n", "L 272.246129 -40.231949 \n", "L 272.665153 -40.213626 \n", "L 273.084176 -40.19536 \n", "L 273.5032 -40.17715 \n", "L 273.922223 -40.158999 \n", "L 274.341246 -40.140907 \n", "L 274.76027 -40.122875 \n", "L 275.179298 -40.104904 \n", "L 275.598321 -40.086995 \n", "L 276.017345 -40.069149 \n", "L 276.436368 -40.051367 \n", "L 276.855392 -40.033649 \n", "L 277.274415 -40.015997 \n", "L 277.693438 -39.998411 \n", "L 278.112462 -39.980892 \n", "L 278.53149 -39.963441 \n", "L 278.950513 -39.94606 \n", "L 279.369537 -39.928747 \n", "L 279.78856 -39.911504 \n", "L 280.207584 -39.894332 \n", "L 280.626607 -39.877232 \n", "L 281.04563 -39.860204 \n", "L 281.464654 -39.843249 \n", "L 281.883682 -39.826366 \n", "L 282.302701 -39.809558 \n", "L 282.721724 -39.792825 \n", "L 283.140747 -39.776167 \n", "L 283.140747 -37.55625 \n", "L 283.140747 -37.55625 \n", "L 282.721724 -37.55625 \n", "L 282.302701 -37.55625 \n", "L 281.883682 -37.55625 \n", "L 281.464654 -37.55625 \n", "L 281.04563 -37.55625 \n", "L 280.626607 -37.55625 \n", "L 280.207584 -37.55625 \n", "L 279.78856 -37.55625 \n", "L 279.369537 -37.55625 \n", "L 278.950513 -37.55625 \n", "L 278.53149 -37.55625 \n", "L 278.112462 -37.55625 \n", "L 277.693438 -37.55625 \n", "L 277.274415 -37.55625 \n", "L 276.855392 -37.55625 \n", "L 276.436368 -37.55625 \n", "L 276.017345 -37.55625 \n", "L 275.598321 -37.55625 \n", "L 275.179298 -37.55625 \n", "L 274.76027 -37.55625 \n", "L 274.341246 -37.55625 \n", "L 273.922223 -37.55625 \n", "L 273.5032 -37.55625 \n", "L 273.084176 -37.55625 \n", "L 272.665153 -37.55625 \n", "L 272.246129 -37.55625 \n", "L 271.827111 -37.55625 \n", "L 271.408083 -37.55625 \n", "L 270.989059 -37.55625 \n", "L 270.570036 -37.55625 \n", "L 270.151013 -37.55625 \n", "L 269.731989 -37.55625 \n", "L 269.312966 -37.55625 \n", "L 268.893942 -37.55625 \n", "L 268.474919 -37.55625 \n", "L 268.055891 -37.55625 \n", "L 267.636867 -37.55625 \n", "L 267.217844 -37.55625 \n", "L 266.79882 -37.55625 \n", "L 266.379797 -37.55625 \n", "L 265.960774 -37.55625 \n", "L 265.54175 -37.55625 \n", "L 265.122732 -37.55625 \n", "L 264.703704 -37.55625 \n", "L 264.28468 -37.55625 \n", "L 263.865657 -37.55625 \n", "L 263.446633 -37.55625 \n", "L 263.02761 -37.55625 \n", "L 262.608587 -37.55625 \n", "L 262.189563 -37.55625 \n", "L 261.77054 -37.55625 \n", "L 261.351512 -37.55625 \n", "L 260.932488 -37.55625 \n", "L 260.513465 -37.55625 \n", "L 260.094441 -37.55625 \n", "L 259.675418 -37.55625 \n", "L 259.256395 -37.55625 \n", "L 258.837371 -37.55625 \n", "L 258.418348 -37.55625 \n", "L 257.99932 -37.55625 \n", "L 257.580296 -37.55625 \n", "L 257.161273 -37.55625 \n", "L 256.742249 -37.55625 \n", "L 256.323226 -37.55625 \n", "L 255.904203 -37.55625 \n", "L 255.485179 -37.55625 \n", "L 255.066161 -37.55625 \n", "L 254.647132 -37.55625 \n", "L 254.228109 -37.55625 \n", "L 253.809086 -37.55625 \n", "L 253.390062 -37.55625 \n", "L 252.971039 -37.55625 \n", "L 252.552016 -37.55625 \n", "L 252.132992 -37.55625 \n", "L 251.713969 -37.55625 \n", "L 251.29494 -37.55625 \n", "L 250.875917 -37.55625 \n", "L 250.456894 -37.55625 \n", "L 250.03787 -37.55625 \n", "L 249.618847 -37.55625 \n", "L 249.199824 -37.55625 \n", "L 248.7808 -37.55625 \n", "L 248.361777 -37.55625 \n", "L 247.942748 -37.55625 \n", "z\n", "\" style=\"stroke: #e377c2; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pfb33315d50)\">\n", "     <use xlink:href=\"#m34741f690e\" x=\"0\" y=\"226.194375\" style=\"fill: #e377c2; fill-opacity: 0.5; stroke: #e377c2; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"PolyCollection_8\">\n", "    <defs>\n", "     <path id=\"m3716da7abd\" d=\"M 283.559771 -37.55625 \n", "L 283.559771 -39.759584 \n", "L 283.978794 -39.743078 \n", "L 284.397817 -39.726647 \n", "L 284.816841 -39.710295 \n", "L 285.235869 -39.69402 \n", "L 285.654893 -39.677823 \n", "L 286.073916 -39.661704 \n", "L 286.492939 -39.645665 \n", "L 286.911963 -39.629705 \n", "L 287.330986 -39.613824 \n", "L 287.750009 -39.598024 \n", "L 288.169033 -39.582303 \n", "L 288.588056 -39.566664 \n", "L 289.00708 -39.551106 \n", "L 289.426103 -39.535628 \n", "L 289.845126 -39.520233 \n", "L 290.26415 -39.50492 \n", "L 290.683173 -39.489688 \n", "L 291.102196 -39.474539 \n", "L 291.52122 -39.459473 \n", "L 291.940243 -39.444489 \n", "L 292.359277 -39.429587 \n", "L 292.7783 -39.41477 \n", "L 293.197323 -39.400036 \n", "L 293.616347 -39.385385 \n", "L 294.03537 -39.370817 \n", "L 294.454393 -39.356333 \n", "L 294.873417 -39.341934 \n", "L 295.29244 -39.327618 \n", "L 295.711464 -39.313386 \n", "L 296.130487 -39.299237 \n", "L 296.54951 -39.285174 \n", "L 296.968534 -39.271194 \n", "L 297.387557 -39.257298 \n", "L 297.806581 -39.243487 \n", "L 298.225604 -39.22976 \n", "L 298.644627 -39.216116 \n", "L 299.063651 -39.202557 \n", "L 299.482674 -39.189082 \n", "L 299.901697 -39.175692 \n", "L 300.320721 -39.162386 \n", "L 300.739744 -39.149163 \n", "L 301.158768 -39.136024 \n", "L 301.577791 -39.12297 \n", "L 301.996814 -39.109998 \n", "L 302.415848 -39.097111 \n", "L 302.834871 -39.084307 \n", "L 303.253894 -39.071587 \n", "L 303.672918 -39.05895 \n", "L 304.091941 -39.046397 \n", "L 304.510965 -39.033926 \n", "L 304.929988 -39.021538 \n", "L 305.349011 -39.009234 \n", "L 305.768035 -38.997011 \n", "L 306.187058 -38.984872 \n", "L 306.606081 -38.972814 \n", "L 307.025105 -38.960838 \n", "L 307.444128 -38.948945 \n", "L 307.863152 -38.937133 \n", "L 308.282175 -38.925403 \n", "L 308.701198 -38.913753 \n", "L 309.120222 -38.902185 \n", "L 309.539245 -38.890698 \n", "L 309.958269 -38.879291 \n", "L 310.377292 -38.867965 \n", "L 310.796315 -38.856718 \n", "L 311.215339 -38.845552 \n", "L 311.634362 -38.834465 \n", "L 312.053385 -38.823457 \n", "L 312.472419 -38.812528 \n", "L 312.891442 -38.801678 \n", "L 313.310466 -38.790907 \n", "L 313.729489 -38.780214 \n", "L 314.148512 -38.769598 \n", "L 314.567536 -38.759061 \n", "L 314.986559 -38.7486 \n", "L 315.405582 -38.738217 \n", "L 315.824606 -38.727911 \n", "L 316.243629 -38.71768 \n", "L 316.662653 -38.707526 \n", "L 317.081676 -38.697448 \n", "L 317.500699 -38.687445 \n", "L 317.919723 -38.677517 \n", "L 318.338746 -38.667664 \n", "L 318.757769 -38.657886 \n", "L 319.176793 -38.648182 \n", "L 319.595816 -38.638551 \n", "L 320.01484 -38.628994 \n", "L 320.433863 -38.61951 \n", "L 320.852886 -38.610098 \n", "L 321.27191 -38.60076 \n", "L 321.690933 -38.591493 \n", "L 322.109957 -38.582298 \n", "L 322.52899 -38.573174 \n", "L 322.948013 -38.564121 \n", "L 323.367037 -38.555139 \n", "L 323.78606 -38.546227 \n", "L 324.205083 -38.537385 \n", "L 324.624107 -38.528612 \n", "L 325.04313 -38.519909 \n", "L 325.462154 -38.511275 \n", "L 325.881177 -38.502708 \n", "L 326.3002 -38.49421 \n", "L 326.719224 -38.48578 \n", "L 327.138247 -38.477417 \n", "L 327.55727 -38.469121 \n", "L 327.976294 -38.460891 \n", "L 328.395317 -38.452728 \n", "L 328.814341 -38.44463 \n", "L 329.233364 -38.436598 \n", "L 329.652387 -38.42863 \n", "L 330.071411 -38.420728 \n", "L 330.490434 -38.412889 \n", "L 330.909457 -38.405115 \n", "L 331.328481 -38.397404 \n", "L 331.747504 -38.389756 \n", "L 332.166528 -38.382171 \n", "L 332.585551 -38.374648 \n", "L 333.004574 -38.367187 \n", "L 333.423598 -38.359788 \n", "L 333.842621 -38.35245 \n", "L 334.261645 -38.345173 \n", "L 334.680668 -38.337956 \n", "L 335.099691 -38.330799 \n", "L 335.518715 -38.323702 \n", "L 335.937748 -38.316663 \n", "L 336.356771 -38.309684 \n", "L 336.775795 -38.302764 \n", "L 337.194818 -38.295901 \n", "L 337.613842 -38.289096 \n", "L 338.032865 -38.282349 \n", "L 338.451888 -38.275658 \n", "L 338.870912 -38.269024 \n", "L 339.289935 -38.262447 \n", "L 339.708958 -38.255925 \n", "L 340.127982 -38.249458 \n", "L 340.547005 -38.243047 \n", "L 340.966029 -38.23669 \n", "L 341.385052 -38.230388 \n", "L 341.804075 -38.224139 \n", "L 342.223099 -38.217944 \n", "L 342.642122 -38.211802 \n", "L 343.061145 -38.205713 \n", "L 343.480169 -38.199677 \n", "L 343.899192 -38.193692 \n", "L 344.318216 -38.187759 \n", "L 344.737239 -38.181877 \n", "L 345.156262 -38.176047 \n", "L 345.575286 -38.170267 \n", "L 345.994319 -38.164537 \n", "L 346.413343 -38.158857 \n", "L 346.832366 -38.153226 \n", "L 347.251389 -38.147645 \n", "L 347.670413 -38.142112 \n", "L 348.089436 -38.136628 \n", "L 348.508459 -38.131191 \n", "L 348.927483 -38.125803 \n", "L 349.346506 -38.120462 \n", "L 349.76553 -38.115167 \n", "L 350.184553 -38.10992 \n", "L 350.603576 -38.104718 \n", "L 351.0226 -38.099563 \n", "L 351.441623 -38.094453 \n", "L 351.860646 -38.089388 \n", "L 352.27967 -38.084368 \n", "L 352.698693 -38.079393 \n", "L 353.117717 -38.074462 \n", "L 353.53674 -38.069575 \n", "L 353.955763 -38.064731 \n", "L 354.374787 -38.05993 \n", "L 354.79381 -38.055173 \n", "L 355.212833 -38.050457 \n", "L 355.631857 -38.045784 \n", "L 356.05089 -38.041153 \n", "L 356.469914 -38.036563 \n", "L 356.888937 -38.032015 \n", "L 357.30796 -38.027507 \n", "L 357.726984 -38.02304 \n", "L 358.146007 -38.018613 \n", "L 358.565031 -38.014226 \n", "L 358.984054 -38.009878 \n", "L 359.403077 -38.00557 \n", "L 359.822101 -38.001301 \n", "L 360.241124 -37.99707 \n", "L 360.660147 -37.992878 \n", "L 361.079171 -37.988723 \n", "L 361.498194 -37.984607 \n", "L 361.917218 -37.980527 \n", "L 362.336241 -37.976485 \n", "L 362.755264 -37.97248 \n", "L 363.174288 -37.968511 \n", "L 363.593311 -37.964578 \n", "L 364.012334 -37.960681 \n", "L 364.431358 -37.95682 \n", "L 364.850381 -37.952994 \n", "L 365.269405 -37.949203 \n", "L 365.688428 -37.945446 \n", "L 366.107461 -37.941725 \n", "L 366.526485 -37.938037 \n", "L 366.945508 -37.934383 \n", "L 367.364531 -37.930763 \n", "L 367.783555 -37.927176 \n", "L 368.202578 -37.923622 \n", "L 368.621602 -37.9201 \n", "L 369.040625 -37.916611 \n", "L 369.040625 -37.55625 \n", "L 369.040625 -37.55625 \n", "L 368.621602 -37.55625 \n", "L 368.202578 -37.55625 \n", "L 367.783555 -37.55625 \n", "L 367.364531 -37.55625 \n", "L 366.945508 -37.55625 \n", "L 366.526485 -37.55625 \n", "L 366.107461 -37.55625 \n", "L 365.688428 -37.55625 \n", "L 365.269405 -37.55625 \n", "L 364.850381 -37.55625 \n", "L 364.431358 -37.55625 \n", "L 364.012334 -37.55625 \n", "L 363.593311 -37.55625 \n", "L 363.174288 -37.55625 \n", "L 362.755264 -37.55625 \n", "L 362.336241 -37.55625 \n", "L 361.917218 -37.55625 \n", "L 361.498194 -37.55625 \n", "L 361.079171 -37.55625 \n", "L 360.660147 -37.55625 \n", "L 360.241124 -37.55625 \n", "L 359.822101 -37.55625 \n", "L 359.403077 -37.55625 \n", "L 358.984054 -37.55625 \n", "L 358.565031 -37.55625 \n", "L 358.146007 -37.55625 \n", "L 357.726984 -37.55625 \n", "L 357.30796 -37.55625 \n", "L 356.888937 -37.55625 \n", "L 356.469914 -37.55625 \n", "L 356.05089 -37.55625 \n", "L 355.631857 -37.55625 \n", "L 355.212833 -37.55625 \n", "L 354.79381 -37.55625 \n", "L 354.374787 -37.55625 \n", "L 353.955763 -37.55625 \n", "L 353.53674 -37.55625 \n", "L 353.117717 -37.55625 \n", "L 352.698693 -37.55625 \n", "L 352.27967 -37.55625 \n", "L 351.860646 -37.55625 \n", "L 351.441623 -37.55625 \n", "L 351.0226 -37.55625 \n", "L 350.603576 -37.55625 \n", "L 350.184553 -37.55625 \n", "L 349.76553 -37.55625 \n", "L 349.346506 -37.55625 \n", "L 348.927483 -37.55625 \n", "L 348.508459 -37.55625 \n", "L 348.089436 -37.55625 \n", "L 347.670413 -37.55625 \n", "L 347.251389 -37.55625 \n", "L 346.832366 -37.55625 \n", "L 346.413343 -37.55625 \n", "L 345.994319 -37.55625 \n", "L 345.575286 -37.55625 \n", "L 345.156262 -37.55625 \n", "L 344.737239 -37.55625 \n", "L 344.318216 -37.55625 \n", "L 343.899192 -37.55625 \n", "L 343.480169 -37.55625 \n", "L 343.061145 -37.55625 \n", "L 342.642122 -37.55625 \n", "L 342.223099 -37.55625 \n", "L 341.804075 -37.55625 \n", "L 341.385052 -37.55625 \n", "L 340.966029 -37.55625 \n", "L 340.547005 -37.55625 \n", "L 340.127982 -37.55625 \n", "L 339.708958 -37.55625 \n", "L 339.289935 -37.55625 \n", "L 338.870912 -37.55625 \n", "L 338.451888 -37.55625 \n", "L 338.032865 -37.55625 \n", "L 337.613842 -37.55625 \n", "L 337.194818 -37.55625 \n", "L 336.775795 -37.55625 \n", "L 336.356771 -37.55625 \n", "L 335.937748 -37.55625 \n", "L 335.518715 -37.55625 \n", "L 335.099691 -37.55625 \n", "L 334.680668 -37.55625 \n", "L 334.261645 -37.55625 \n", "L 333.842621 -37.55625 \n", "L 333.423598 -37.55625 \n", "L 333.004574 -37.55625 \n", "L 332.585551 -37.55625 \n", "L 332.166528 -37.55625 \n", "L 331.747504 -37.55625 \n", "L 331.328481 -37.55625 \n", "L 330.909457 -37.55625 \n", "L 330.490434 -37.55625 \n", "L 330.071411 -37.55625 \n", "L 329.652387 -37.55625 \n", "L 329.233364 -37.55625 \n", "L 328.814341 -37.55625 \n", "L 328.395317 -37.55625 \n", "L 327.976294 -37.55625 \n", "L 327.55727 -37.55625 \n", "L 327.138247 -37.55625 \n", "L 326.719224 -37.55625 \n", "L 326.3002 -37.55625 \n", "L 325.881177 -37.55625 \n", "L 325.462154 -37.55625 \n", "L 325.04313 -37.55625 \n", "L 324.624107 -37.55625 \n", "L 324.205083 -37.55625 \n", "L 323.78606 -37.55625 \n", "L 323.367037 -37.55625 \n", "L 322.948013 -37.55625 \n", "L 322.52899 -37.55625 \n", "L 322.109957 -37.55625 \n", "L 321.690933 -37.55625 \n", "L 321.27191 -37.55625 \n", "L 320.852886 -37.55625 \n", "L 320.433863 -37.55625 \n", "L 320.01484 -37.55625 \n", "L 319.595816 -37.55625 \n", "L 319.176793 -37.55625 \n", "L 318.757769 -37.55625 \n", "L 318.338746 -37.55625 \n", "L 317.919723 -37.55625 \n", "L 317.500699 -37.55625 \n", "L 317.081676 -37.55625 \n", "L 316.662653 -37.55625 \n", "L 316.243629 -37.55625 \n", "L 315.824606 -37.55625 \n", "L 315.405582 -37.55625 \n", "L 314.986559 -37.55625 \n", "L 314.567536 -37.55625 \n", "L 314.148512 -37.55625 \n", "L 313.729489 -37.55625 \n", "L 313.310466 -37.55625 \n", "L 312.891442 -37.55625 \n", "L 312.472419 -37.55625 \n", "L 312.053385 -37.55625 \n", "L 311.634362 -37.55625 \n", "L 311.215339 -37.55625 \n", "L 310.796315 -37.55625 \n", "L 310.377292 -37.55625 \n", "L 309.958269 -37.55625 \n", "L 309.539245 -37.55625 \n", "L 309.120222 -37.55625 \n", "L 308.701198 -37.55625 \n", "L 308.282175 -37.55625 \n", "L 307.863152 -37.55625 \n", "L 307.444128 -37.55625 \n", "L 307.025105 -37.55625 \n", "L 306.606081 -37.55625 \n", "L 306.187058 -37.55625 \n", "L 305.768035 -37.55625 \n", "L 305.349011 -37.55625 \n", "L 304.929988 -37.55625 \n", "L 304.510965 -37.55625 \n", "L 304.091941 -37.55625 \n", "L 303.672918 -37.55625 \n", "L 303.253894 -37.55625 \n", "L 302.834871 -37.55625 \n", "L 302.415848 -37.55625 \n", "L 301.996814 -37.55625 \n", "L 301.577791 -37.55625 \n", "L 301.158768 -37.55625 \n", "L 300.739744 -37.55625 \n", "L 300.320721 -37.55625 \n", "L 299.901697 -37.55625 \n", "L 299.482674 -37.55625 \n", "L 299.063651 -37.55625 \n", "L 298.644627 -37.55625 \n", "L 298.225604 -37.55625 \n", "L 297.806581 -37.55625 \n", "L 297.387557 -37.55625 \n", "L 296.968534 -37.55625 \n", "L 296.54951 -37.55625 \n", "L 296.130487 -37.55625 \n", "L 295.711464 -37.55625 \n", "L 295.29244 -37.55625 \n", "L 294.873417 -37.55625 \n", "L 294.454393 -37.55625 \n", "L 294.03537 -37.55625 \n", "L 293.616347 -37.55625 \n", "L 293.197323 -37.55625 \n", "L 292.7783 -37.55625 \n", "L 292.359277 -37.55625 \n", "L 291.940243 -37.55625 \n", "L 291.52122 -37.55625 \n", "L 291.102196 -37.55625 \n", "L 290.683173 -37.55625 \n", "L 290.26415 -37.55625 \n", "L 289.845126 -37.55625 \n", "L 289.426103 -37.55625 \n", "L 289.00708 -37.55625 \n", "L 288.588056 -37.55625 \n", "L 288.169033 -37.55625 \n", "L 287.750009 -37.55625 \n", "L 287.330986 -37.55625 \n", "L 286.911963 -37.55625 \n", "L 286.492939 -37.55625 \n", "L 286.073916 -37.55625 \n", "L 285.654893 -37.55625 \n", "L 285.235869 -37.55625 \n", "L 284.816841 -37.55625 \n", "L 284.397817 -37.55625 \n", "L 283.978794 -37.55625 \n", "L 283.559771 -37.55625 \n", "z\n", "\" style=\"stroke: #7f7f7f; stroke-opacity: 0.5\"/>\n", "    </defs>\n", "    <g clip-path=\"url(#pfb33315d50)\">\n", "     <use xlink:href=\"#m3716da7abd\" x=\"0\" y=\"226.194375\" style=\"fill: #7f7f7f; fill-opacity: 0.5; stroke: #7f7f7f; stroke-opacity: 0.5\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"line2d_1\">\n", "    <path d=\"M 34.240625 188.638125 \n", "L 34.240625 186.496696 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #1f77b4; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_2\">\n", "    <path d=\"M 120.140498 188.638125 \n", "L 120.140498 175.418124 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #1f77b4; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_3\">\n", "    <path d=\"M 120.559521 188.638125 \n", "L 120.559521 153.119465 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #ff7f0e; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_4\">\n", "    <path d=\"M 155.75752 188.638125 \n", "L 155.75752 128.060121 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #ff7f0e; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_5\">\n", "    <path d=\"M 156.176543 188.638125 \n", "L 156.176543 66.876491 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #2ca02c; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_6\">\n", "    <path d=\"M 180.060901 188.638125 \n", "L 180.060901 37.438125 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #2ca02c; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_7\">\n", "    <path d=\"M 180.479924 188.638125 \n", "L 180.479924 112.847409 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #d62728; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_8\">\n", "    <path d=\"M 201.431113 188.638125 \n", "L 201.431113 107.813393 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #d62728; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_9\">\n", "    <path d=\"M 201.850137 188.638125 \n", "L 201.850137 158.328092 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #9467bd; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_10\">\n", "    <path d=\"M 223.220349 188.638125 \n", "L 223.220349 160.216606 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #9467bd; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_11\">\n", "    <path d=\"M 223.639372 188.638125 \n", "L 223.639372 179.188126 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #8c564b; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_12\">\n", "    <path d=\"M 247.523725 188.638125 \n", "L 247.523725 181.028022 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #8c564b; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_13\">\n", "    <path d=\"M 247.942748 188.638125 \n", "L 247.942748 184.852 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #e377c2; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_14\">\n", "    <path d=\"M 283.140747 188.638125 \n", "L 283.140747 186.418208 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #e377c2; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_15\">\n", "    <path d=\"M 283.559771 188.638125 \n", "L 283.559771 186.434791 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #7f7f7f; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_16\">\n", "    <path d=\"M 369.040625 188.638125 \n", "L 369.040625 188.277764 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #7f7f7f; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"line2d_17\">\n", "    <path d=\"M 34.240625 186.496696 \n", "L 47.649383 185.728708 \n", "L 59.801071 184.813475 \n", "L 70.695689 183.772503 \n", "L 80.75226 182.590899 \n", "L 90.389808 181.230184 \n", "L 99.608332 179.697245 \n", "L 108.407833 178.00992 \n", "L 117.207334 176.101637 \n", "L 120.140498 175.418124 \n", "L 120.559521 153.119465 \n", "L 126.844877 148.999741 \n", "L 133.968284 144.050133 \n", "L 142.76778 137.644128 \n", "L 155.75752 128.060121 \n", "L 156.176543 66.876491 \n", "L 162.042878 58.598824 \n", "L 166.65214 52.453728 \n", "L 170.842377 47.241436 \n", "L 174.61359 42.92512 \n", "L 177.965782 39.435311 \n", "L 180.060901 37.438125 \n", "L 180.479924 112.847409 \n", "L 184.251139 111.273603 \n", "L 187.603328 110.102977 \n", "L 190.955519 109.162099 \n", "L 194.307709 108.462547 \n", "L 197.659899 108.013104 \n", "L 200.593065 107.829552 \n", "L 201.431113 107.813393 \n", "L 201.850137 158.328092 \n", "L 206.878422 158.436948 \n", "L 211.906707 158.760399 \n", "L 217.354017 159.342234 \n", "L 222.801326 160.146244 \n", "L 223.220349 160.216606 \n", "L 223.639372 179.188126 \n", "L 234.533993 179.928118 \n", "L 247.523725 181.028022 \n", "L 247.942748 184.852 \n", "L 293.616347 186.80899 \n", "L 315.405582 187.456158 \n", "L 338.870912 187.925351 \n", "L 367.364531 188.263612 \n", "L 369.040625 188.277764 \n", "L 369.040625 188.277764 \n", "\" clip-path=\"url(#pfb33315d50)\" style=\"fill: none; stroke: #000000; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"patch_3\">\n", "    <path d=\"M 34.240625 188.638125 \n", "L 34.240625 22.318125 \n", "\" style=\"fill: none; stroke: #262626; stroke-width: 0.8; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_4\">\n", "    <path d=\"M 369.040625 188.638125 \n", "L 369.040625 22.318125 \n", "\" style=\"fill: none; stroke: #262626; stroke-width: 0.8; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_5\">\n", "    <path d=\"M 34.240625 188.638125 \n", "L 369.040625 188.638125 \n", "\" style=\"fill: none; stroke: #262626; stroke-width: 0.8; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_6\">\n", "    <path d=\"M 34.240625 22.318125 \n", "L 369.040625 22.318125 \n", "\" style=\"fill: none; stroke: #262626; stroke-width: 0.8; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"text_19\">\n", "    <!-- Dequantization distribution for 8 discrete values -->\n", "    <g style=\"fill: #262626\" transform=\"translate(56.623438 16.318125)scale(0.12 -0.12)\">\n", "     <defs>\n", "      <path id=\"DejaVuSans-44\" d=\"M 1259 4147 \n", "L 1259 519 \n", "L 2022 519 \n", "Q 2988 519 3436 956 \n", "Q 3884 1394 3884 2338 \n", "Q 3884 3275 3436 3711 \n", "Q 2988 4147 2022 4147 \n", "L 1259 4147 \n", "z\n", "M 628 4666 \n", "L 1925 4666 \n", "Q 3281 4666 3915 4102 \n", "Q 4550 3538 4550 2338 \n", "Q 4550 1131 3912 565 \n", "Q 3275 0 1925 0 \n", "L 628 0 \n", "L 628 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-65\" d=\"M 3597 1894 \n", "L 3597 1613 \n", "L 953 1613 \n", "Q 991 1019 1311 708 \n", "Q 1631 397 2203 397 \n", "Q 2534 397 2845 478 \n", "Q 3156 559 3463 722 \n", "L 3463 178 \n", "Q 3153 47 2828 -22 \n", "Q 2503 -91 2169 -91 \n", "Q 1331 -91 842 396 \n", "Q 353 884 353 1716 \n", "Q 353 2575 817 3079 \n", "Q 1281 3584 2069 3584 \n", "Q 2775 3584 3186 3129 \n", "Q 3597 2675 3597 1894 \n", "z\n", "M 3022 2063 \n", "Q 3016 2534 2758 2815 \n", "Q 2500 3097 2075 3097 \n", "Q 1594 3097 1305 2825 \n", "Q 1016 2553 972 2059 \n", "L 3022 2063 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-71\" d=\"M 947 1747 \n", "Q 947 1113 1208 752 \n", "Q 1469 391 1925 391 \n", "Q 2381 391 2643 752 \n", "Q 2906 1113 2906 1747 \n", "Q 2906 2381 2643 2742 \n", "Q 2381 3103 1925 3103 \n", "Q 1469 3103 1208 2742 \n", "Q 947 2381 947 1747 \n", "z\n", "M 2906 525 \n", "Q 2725 213 2448 61 \n", "Q 2172 -91 1784 -91 \n", "Q 1150 -91 751 415 \n", "Q 353 922 353 1747 \n", "Q 353 2572 751 3078 \n", "Q 1150 3584 1784 3584 \n", "Q 2172 3584 2448 3432 \n", "Q 2725 3281 2906 2969 \n", "L 2906 3500 \n", "L 3481 3500 \n", "L 3481 -1331 \n", "L 2906 -1331 \n", "L 2906 525 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-75\" d=\"M 544 1381 \n", "L 544 3500 \n", "L 1119 3500 \n", "L 1119 1403 \n", "Q 1119 906 1312 657 \n", "Q 1506 409 1894 409 \n", "Q 2359 409 2629 706 \n", "Q 2900 1003 2900 1516 \n", "L 2900 3500 \n", "L 3475 3500 \n", "L 3475 0 \n", "L 2900 0 \n", "L 2900 538 \n", "Q 2691 219 2414 64 \n", "Q 2138 -91 1772 -91 \n", "Q 1169 -91 856 284 \n", "Q 544 659 544 1381 \n", "z\n", "M 1991 3584 \n", "L 1991 3584 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6e\" d=\"M 3513 2113 \n", "L 3513 0 \n", "L 2938 0 \n", "L 2938 2094 \n", "Q 2938 2591 2744 2837 \n", "Q 2550 3084 2163 3084 \n", "Q 1697 3084 1428 2787 \n", "Q 1159 2491 1159 1978 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 3500 \n", "L 1159 3500 \n", "L 1159 2956 \n", "Q 1366 3272 1645 3428 \n", "Q 1925 3584 2291 3584 \n", "Q 2894 3584 3203 3211 \n", "Q 3513 2838 3513 2113 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-20\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-64\" d=\"M 2906 2969 \n", "L 2906 4863 \n", "L 3481 4863 \n", "L 3481 0 \n", "L 2906 0 \n", "L 2906 525 \n", "Q 2725 213 2448 61 \n", "Q 2172 -91 1784 -91 \n", "Q 1150 -91 751 415 \n", "Q 353 922 353 1747 \n", "Q 353 2572 751 3078 \n", "Q 1150 3584 1784 3584 \n", "Q 2172 3584 2448 3432 \n", "Q 2725 3281 2906 2969 \n", "z\n", "M 947 1747 \n", "Q 947 1113 1208 752 \n", "Q 1469 391 1925 391 \n", "Q 2381 391 2643 752 \n", "Q 2906 1113 2906 1747 \n", "Q 2906 2381 2643 2742 \n", "Q 2381 3103 1925 3103 \n", "Q 1469 3103 1208 2742 \n", "Q 947 2381 947 1747 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-73\" d=\"M 2834 3397 \n", "L 2834 2853 \n", "Q 2591 2978 2328 3040 \n", "Q 2066 3103 1784 3103 \n", "Q 1356 3103 1142 2972 \n", "Q 928 2841 928 2578 \n", "Q 928 2378 1081 2264 \n", "Q 1234 2150 1697 2047 \n", "L 1894 2003 \n", "Q 2506 1872 2764 1633 \n", "Q 3022 1394 3022 966 \n", "Q 3022 478 2636 193 \n", "Q 2250 -91 1575 -91 \n", "Q 1294 -91 989 -36 \n", "Q 684 19 347 128 \n", "L 347 722 \n", "Q 666 556 975 473 \n", "Q 1284 391 1588 391 \n", "Q 1994 391 2212 530 \n", "Q 2431 669 2431 922 \n", "Q 2431 1156 2273 1281 \n", "Q 2116 1406 1581 1522 \n", "L 1381 1569 \n", "Q 847 1681 609 1914 \n", "Q 372 2147 372 2553 \n", "Q 372 3047 722 3315 \n", "Q 1072 3584 1716 3584 \n", "Q 2034 3584 2315 3537 \n", "Q 2597 3491 2834 3397 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-66\" d=\"M 2375 4863 \n", "L 2375 4384 \n", "L 1825 4384 \n", "Q 1516 4384 1395 4259 \n", "Q 1275 4134 1275 3809 \n", "L 1275 3500 \n", "L 2222 3500 \n", "L 2222 3053 \n", "L 1275 3053 \n", "L 1275 0 \n", "L 697 0 \n", "L 697 3053 \n", "L 147 3053 \n", "L 147 3500 \n", "L 697 3500 \n", "L 697 3744 \n", "Q 697 4328 969 4595 \n", "Q 1241 4863 1831 4863 \n", "L 2375 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-38\" d=\"M 2034 2216 \n", "Q 1584 2216 1326 1975 \n", "Q 1069 1734 1069 1313 \n", "Q 1069 891 1326 650 \n", "Q 1584 409 2034 409 \n", "Q 2484 409 2743 651 \n", "Q 3003 894 3003 1313 \n", "Q 3003 1734 2745 1975 \n", "Q 2488 2216 2034 2216 \n", "z\n", "M 1403 2484 \n", "Q 997 2584 770 2862 \n", "Q 544 3141 544 3541 \n", "Q 544 4100 942 4425 \n", "Q 1341 4750 2034 4750 \n", "Q 2731 4750 3128 4425 \n", "Q 3525 4100 3525 3541 \n", "Q 3525 3141 3298 2862 \n", "Q 3072 2584 2669 2484 \n", "Q 3125 2378 3379 2068 \n", "Q 3634 1759 3634 1313 \n", "Q 3634 634 3220 271 \n", "Q 2806 -91 2034 -91 \n", "Q 1263 -91 848 271 \n", "Q 434 634 434 1313 \n", "Q 434 1759 690 2068 \n", "Q 947 2378 1403 2484 \n", "z\n", "M 1172 3481 \n", "Q 1172 3119 1398 2916 \n", "Q 1625 2713 2034 2713 \n", "Q 2441 2713 2670 2916 \n", "Q 2900 3119 2900 3481 \n", "Q 2900 3844 2670 4047 \n", "Q 2441 4250 2034 4250 \n", "Q 1625 4250 1398 4047 \n", "Q 1172 3844 1172 3481 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-63\" d=\"M 3122 3366 \n", "L 3122 2828 \n", "Q 2878 2963 2633 3030 \n", "Q 2388 3097 2138 3097 \n", "Q 1578 3097 1268 2742 \n", "Q 959 2388 959 1747 \n", "Q 959 1106 1268 751 \n", "Q 1578 397 2138 397 \n", "Q 2388 397 2633 464 \n", "Q 2878 531 3122 666 \n", "L 3122 134 \n", "Q 2881 22 2623 -34 \n", "Q 2366 -91 2075 -91 \n", "Q 1284 -91 818 406 \n", "Q 353 903 353 1747 \n", "Q 353 2603 823 3093 \n", "Q 1294 3584 2113 3584 \n", "Q 2378 3584 2631 3529 \n", "Q 2884 3475 3122 3366 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-76\" d=\"M 191 3500 \n", "L 800 3500 \n", "L 1894 563 \n", "L 2988 3500 \n", "L 3597 3500 \n", "L 2284 0 \n", "L 1503 0 \n", "L 191 3500 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "     </defs>\n", "     <use xlink:href=\"#DejaVuSans-44\"/>\n", "     <use xlink:href=\"#DejaVuSans-65\" x=\"77.001953\"/>\n", "     <use xlink:href=\"#DejaVuSans-71\" x=\"138.525391\"/>\n", "     <use xlink:href=\"#DejaVuSans-75\" x=\"202.001953\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"265.380859\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"326.660156\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"390.039062\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"429.248047\"/>\n", "     <use xlink:href=\"#DejaVuSans-7a\" x=\"457.03125\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"509.521484\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"570.800781\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"610.009766\"/>\n", "     <use xlink:href=\"#DejaVuSans-6f\" x=\"637.792969\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"698.974609\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"762.353516\"/>\n", "     <use xlink:href=\"#DejaVuSans-64\" x=\"794.140625\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"857.617188\"/>\n", "     <use xlink:href=\"#DejaVuSans-73\" x=\"885.400391\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"937.5\"/>\n", "     <use xlink:href=\"#DejaVuSans-72\" x=\"976.708984\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"1017.822266\"/>\n", "     <use xlink:href=\"#DejaVuSans-62\" x=\"1045.605469\"/>\n", "     <use xlink:href=\"#DejaVuSans-75\" x=\"1109.082031\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"1172.460938\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"1211.669922\"/>\n", "     <use xlink:href=\"#DejaVuSans-6f\" x=\"1239.453125\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"1300.634766\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"1364.013672\"/>\n", "     <use xlink:href=\"#DejaVuSans-66\" x=\"1395.800781\"/>\n", "     <use xlink:href=\"#DejaVuSans-6f\" x=\"1431.005859\"/>\n", "     <use xlink:href=\"#DejaVuSans-72\" x=\"1492.1875\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"1533.300781\"/>\n", "     <use xlink:href=\"#DejaVuSans-38\" x=\"1565.087891\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"1628.710938\"/>\n", "     <use xlink:href=\"#DejaVuSans-64\" x=\"1660.498047\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"1723.974609\"/>\n", "     <use xlink:href=\"#DejaVuSans-73\" x=\"1751.757812\"/>\n", "     <use xlink:href=\"#DejaVuSans-63\" x=\"1803.857422\"/>\n", "     <use xlink:href=\"#DejaVuSans-72\" x=\"1858.837891\"/>\n", "     <use xlink:href=\"#DejaVuSans-65\" x=\"1897.701172\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"1959.224609\"/>\n", "     <use xlink:href=\"#DejaVuSans-65\" x=\"1998.433594\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"2059.957031\"/>\n", "     <use xlink:href=\"#DejaVuSans-76\" x=\"2091.744141\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"2150.923828\"/>\n", "     <use xlink:href=\"#DejaVuSans-6c\" x=\"2212.203125\"/>\n", "     <use xlink:href=\"#DejaVuSans-75\" x=\"2239.986328\"/>\n", "     <use xlink:href=\"#DejaVuSans-65\" x=\"2303.365234\"/>\n", "     <use xlink:href=\"#DejaVuSans-73\" x=\"2364.888672\"/>\n", "    </g>\n", "   </g>\n", "   <g id=\"legend_1\">\n", "    <g id=\"patch_7\">\n", "     <path d=\"M 323.678125 147.743125 \n", "L 362.040625 147.743125 \n", "Q 364.040625 147.743125 364.040625 145.743125 \n", "L 364.040625 29.318125 \n", "Q 364.040625 27.318125 362.040625 27.318125 \n", "L 323.678125 27.318125 \n", "Q 321.678125 27.318125 321.678125 29.318125 \n", "L 321.678125 145.743125 \n", "Q 321.678125 147.743125 323.678125 147.743125 \n", "z\n", "\" style=\"fill: #ffffff; opacity: 0.8; stroke: #cccccc; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"patch_8\">\n", "     <path d=\"M 325.678125 38.916562 \n", "L 345.678125 38.916562 \n", "L 345.678125 31.916562 \n", "L 325.678125 31.916562 \n", "z\n", "\" style=\"fill: #1f77b4; fill-opacity: 0.5; stroke: #1f77b4; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_20\">\n", "     <!-- 0 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(353.678125 38.916562)scale(0.1 -0.1)\">\n", "      <use xlink:href=\"#DejaVuSans-30\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_9\">\n", "     <path d=\"M 325.678125 53.594688 \n", "L 345.678125 53.594688 \n", "L 345.678125 46.594688 \n", "L 325.678125 46.594688 \n", "z\n", "\" style=\"fill: #ff7f0e; fill-opacity: 0.5; stroke: #ff7f0e; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_21\">\n", "     <!-- 1 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(353.678125 53.594688)scale(0.1 -0.1)\">\n", "      <use xlink:href=\"#DejaVuSans-31\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_10\">\n", "     <path d=\"M 325.678125 68.272812 \n", "L 345.678125 68.272812 \n", "L 345.678125 61.272812 \n", "L 325.678125 61.272812 \n", "z\n", "\" style=\"fill: #2ca02c; fill-opacity: 0.5; stroke: #2ca02c; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_22\">\n", "     <!-- 2 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(353.678125 68.272812)scale(0.1 -0.1)\">\n", "      <use xlink:href=\"#DejaVuSans-32\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_11\">\n", "     <path d=\"M 325.678125 82.950938 \n", "L 345.678125 82.950938 \n", "L 345.678125 75.950938 \n", "L 325.678125 75.950938 \n", "z\n", "\" style=\"fill: #d62728; fill-opacity: 0.5; stroke: #d62728; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_23\">\n", "     <!-- 3 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(353.678125 82.950938)scale(0.1 -0.1)\">\n", "      <use xlink:href=\"#DejaVuSans-33\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_12\">\n", "     <path d=\"M 325.678125 97.629063 \n", "L 345.678125 97.629063 \n", "L 345.678125 90.629063 \n", "L 325.678125 90.629063 \n", "z\n", "\" style=\"fill: #9467bd; fill-opacity: 0.5; stroke: #9467bd; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_24\">\n", "     <!-- 4 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(353.678125 97.629063)scale(0.1 -0.1)\">\n", "      <use xlink:href=\"#DejaVuSans-34\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_13\">\n", "     <path d=\"M 325.678125 112.307187 \n", "L 345.678125 112.307187 \n", "L 345.678125 105.307187 \n", "L 325.678125 105.307187 \n", "z\n", "\" style=\"fill: #8c564b; fill-opacity: 0.5; stroke: #8c564b; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_25\">\n", "     <!-- 5 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(353.678125 112.307187)scale(0.1 -0.1)\">\n", "      <use xlink:href=\"#DejaVuSans-35\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_14\">\n", "     <path d=\"M 325.678125 126.985312 \n", "L 345.678125 126.985312 \n", "L 345.678125 119.985312 \n", "L 325.678125 119.985312 \n", "z\n", "\" style=\"fill: #e377c2; fill-opacity: 0.5; stroke: #e377c2; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_26\">\n", "     <!-- 6 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(353.678125 126.985312)scale(0.1 -0.1)\">\n", "      <use xlink:href=\"#DejaVuSans-36\"/>\n", "     </g>\n", "    </g>\n", "    <g id=\"patch_15\">\n", "     <path d=\"M 325.678125 141.663437 \n", "L 345.678125 141.663437 \n", "L 345.678125 134.663437 \n", "L 325.678125 134.663437 \n", "z\n", "\" style=\"fill: #7f7f7f; fill-opacity: 0.5; stroke: #7f7f7f; stroke-opacity: 0.5; stroke-linejoin: miter\"/>\n", "    </g>\n", "    <g id=\"text_27\">\n", "     <!-- 7 -->\n", "     <g style=\"fill: #262626\" transform=\"translate(353.678125 141.663437)scale(0.1 -0.1)\">\n", "      <defs>\n", "       <path id=\"DejaVuSans-37\" d=\"M 525 4666 \n", "L 3525 4666 \n", "L 3525 4397 \n", "L 1831 0 \n", "L 1172 0 \n", "L 2766 4134 \n", "L 525 4134 \n", "L 525 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      </defs>\n", "      <use xlink:href=\"#DejaVuSans-37\"/>\n", "     </g>\n", "    </g>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"pfb33315d50\">\n", "   <rect x=\"34.240625\" y=\"22.318125\" width=\"334.8\" height=\"166.32\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 600x300 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["visualize_dequantization(quants=8, prior=np.array([0.075, 0.2, 0.4, 0.2, 0.075, 0.025, 0.0125, 0.0125]))"]}, {"cell_type": "markdown", "id": "717c85bd", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.016286, "end_time": "2022-05-12T12:19:10.334098", "exception": false, "start_time": "2022-05-12T12:19:10.317812", "status": "completed"}, "tags": []}, "source": ["Transforming such a probability into a Gaussian is a difficult task, especially with such hard borders.\n", "Dequantization has therefore been extended to more sophisticated, learnable distributions beyond uniform in a variational framework.\n", "In particular, if we remember the learning objective\n", "$\\log p(x) = \\log \\mathbb{E}_{u}\\left[\\frac{p(x+u)}{q(u|x)} \\right]$,\n", "the uniform distribution can be replaced by a learned distribution $q_{\\theta}(u|x)$ with support over $u\\in[0,1)^D$.\n", "This approach is called Variational Dequantization and has been proposed by Ho et al.\n", "[3].\n", "How can we learn such a distribution?\n", "We can use a second normalizing flow that takes $x$ as external input and learns a flexible distribution over $u$.\n", "To ensure a support over $[0,1)^D$, we can apply a sigmoid activation function as final flow transformation.\n", "\n", "Inheriting the original dequantization class, we can implement variational dequantization as follows:"]}, {"cell_type": "code", "execution_count": 11, "id": "ea583bfb", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:10.370526Z", "iopub.status.busy": "2022-05-12T12:19:10.370052Z", "iopub.status.idle": "2022-05-12T12:19:10.376986Z", "shell.execute_reply": "2022-05-12T12:19:10.376278Z"}, "lines_to_next_cell": 2, "papermill": {"duration": 0.027603, "end_time": "2022-05-12T12:19:10.378528", "exception": false, "start_time": "2022-05-12T12:19:10.350925", "status": "completed"}, "tags": []}, "outputs": [], "source": ["class VariationalDequantization(Dequantization):\n", "    def __init__(self, var_flows, alpha=1e-5):\n", "        \"\"\"\n", "        Args:\n", "            var_flows: A list of flow transformations to use for modeling q(u|x)\n", "            alpha: Small constant, see Dequantization for details\n", "        \"\"\"\n", "        super().__init__(alpha=alpha)\n", "        self.flows = nn.ModuleList(var_flows)\n", "\n", "    def dequant(self, z, ldj):\n", "        z = z.to(torch.float32)\n", "        img = (z / 255.0) * 2 - 1  # We condition the flows on x, i.e. the original image\n", "\n", "        # Prior of u is a uniform distribution as before\n", "        # As most flow transformations are defined on [-infinity,+infinity], we apply an inverse sigmoid first.\n", "        deq_noise = torch.rand_like(z).detach()\n", "        deq_noise, ldj = self.sigmoid(deq_noise, ldj, reverse=True)\n", "        for flow in self.flows:\n", "            deq_noise, ldj = flow(deq_noise, ldj, reverse=False, orig_img=img)\n", "        deq_noise, ldj = self.sigmoid(deq_noise, ldj, reverse=False)\n", "\n", "        # After the flows, apply u as in standard dequantization\n", "        z = (z + deq_noise) / 256.0\n", "        ldj -= np.log(256.0) * np.prod(z.shape[1:])\n", "        return z, ldj"]}, {"cell_type": "markdown", "id": "97d1bdd1", "metadata": {"papermill": {"duration": 0.01692, "end_time": "2022-05-12T12:19:10.412762", "exception": false, "start_time": "2022-05-12T12:19:10.395842", "status": "completed"}, "tags": []}, "source": ["Variational dequantization can be used as a substitute for dequantization.\n", "We will compare dequantization and variational dequantization in later experiments."]}, {"cell_type": "markdown", "id": "3944ac74", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.017196, "end_time": "2022-05-12T12:19:10.446855", "exception": false, "start_time": "2022-05-12T12:19:10.429659", "status": "completed"}, "tags": []}, "source": ["### Coupling layers\n", "\n", "<div class=\"center-wrapper\"><div class=\"video-wrapper\"><iframe src=\"https://www.youtube.com/embed/YoAWiaEt41Y\" title=\"YouTube video player\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture\" allowfullscreen></iframe></div></div>\n", "\n", "Next, we look at possible transformations to apply inside the flow.\n", "A recent popular flow layer, which works well in combination with deep neural networks,\n", "is the coupling layer introduced by Dinh et al.\n", "[1].\n", "The input $z$ is arbitrarily split into two parts, $z_{1:j}$ and $z_{j+1:d}$, of which the first remains unchanged by the flow.\n", "Yet, $z_{1:j}$ is used to parameterize the transformation for the second part, $z_{j+1:d}$.\n", "Various transformations have been proposed in recent time [3,4], but here we will settle for the simplest and most efficient one: affine coupling.\n", "In this coupling layer, we apply an affine transformation by shifting the input by a bias $\\mu$ and scale it by $\\sigma$.\n", "In other words, our transformation looks as follows:\n", "\n", "$$z'_{j+1:d} = \\mu_{\\theta}(z_{1:j}) + \\sigma_{\\theta}(z_{1:j}) \\odot z_{j+1:d}$$\n", "\n", "The functions $\\mu$ and $\\sigma$ are implemented as a shared neural network,\n", "and the sum and multiplication are performed element-wise.\n", "The LDJ is thereby the sum of the logs of the scaling factors: $\\sum_i \\left[\\log \\sigma_{\\theta}(z_{1:j})\\right]_i$.\n", "Inverting the layer can as simply be done as subtracting the bias and dividing by the scale:\n", "\n", "$$z_{j+1:d} = \\left(z'_{j+1:d} - \\mu_{\\theta}(z_{1:j})\\right) / \\sigma_{\\theta}(z_{1:j})$$\n", "\n", "We can also visualize the coupling layer in form of a computation graph,\n", "where $z_1$ represents $z_{1:j}$, and $z_2$ represents $z_{j+1:d}$:\n", "\n", "<center width=\"100%\" style=\"padding: 10px\"><img src=\"https://github.com/PyTorchLightning/lightning-tutorials/raw/main/course_UvA-DL/09-normalizing-flows/coupling_flow.svg\" width=\"450px\"></center>\n", "\n", "In our implementation, we will realize the splitting of variables as masking.\n", "The variables to be transformed, $z_{j+1:d}$, are masked when passing $z$ to the shared network to predict the transformation parameters.\n", "When applying the transformation, we mask the parameters for $z_{1:j}$\n", "so that we have an identity operation for those variables:"]}, {"cell_type": "code", "execution_count": 12, "id": "de9d1fc7", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:10.482629Z", "iopub.status.busy": "2022-05-12T12:19:10.481896Z", "iopub.status.idle": "2022-05-12T12:19:10.489980Z", "shell.execute_reply": "2022-05-12T12:19:10.489308Z"}, "papermill": {"duration": 0.027711, "end_time": "2022-05-12T12:19:10.491507", "exception": false, "start_time": "2022-05-12T12:19:10.463796", "status": "completed"}, "tags": []}, "outputs": [], "source": ["class CouplingLayer(nn.Module):\n", "    def __init__(self, network, mask, c_in):\n", "        \"\"\"Coupling layer inside a normalizing flow.\n", "\n", "        Args:\n", "            network: A PyTorch nn.Module constituting the deep neural network for mu and sigma.\n", "                      Output shape should be twice the channel size as the input.\n", "            mask: Binary mask (0 or 1) where 0 denotes that the element should be transformed,\n", "                   while 1 means the latent will be used as input to the NN.\n", "            c_in: Number of input channels\n", "        \"\"\"\n", "        super().__init__()\n", "        self.network = network\n", "        self.scaling_factor = nn.Parameter(torch.zeros(c_in))\n", "        # Register mask as buffer as it is a tensor which is not a parameter,\n", "        # but should be part of the modules state.\n", "        self.register_buffer(\"mask\", mask)\n", "\n", "    def forward(self, z, ldj, reverse=False, orig_img=None):\n", "        \"\"\"\n", "        Args:\n", "            z: Latent input to the flow\n", "            ldj: The current ldj of the previous flows.\n", "                  The ldj of this layer will be added to this tensor.\n", "            reverse: If True, we apply the inverse of the layer.\n", "            orig_img (optional): Only needed in VarDeq. Allows external\n", "                                  input to condition the flow on (e.g. original image)\n", "        \"\"\"\n", "        # Apply network to masked input\n", "        z_in = z * self.mask\n", "        if orig_img is None:\n", "            nn_out = self.network(z_in)\n", "        else:\n", "            nn_out = self.network(torch.cat([z_in, orig_img], dim=1))\n", "        s, t = nn_out.chunk(2, dim=1)\n", "\n", "        # Stabilize scaling output\n", "        s_fac = self.scaling_factor.exp().view(1, -1, 1, 1)\n", "        s = torch.tanh(s / s_fac) * s_fac\n", "\n", "        # Mask outputs (only transform the second part)\n", "        s = s * (1 - self.mask)\n", "        t = t * (1 - self.mask)\n", "\n", "        # Affine transformation\n", "        if not reverse:\n", "            # Whether we first shift and then scale, or the other way round,\n", "            # is a design choice, and usually does not have a big impact\n", "            z = (z + t) * torch.exp(s)\n", "            ldj += s.sum(dim=[1, 2, 3])\n", "        else:\n", "            z = (z * torch.exp(-s)) - t\n", "            ldj -= s.sum(dim=[1, 2, 3])\n", "\n", "        return z, ldj"]}, {"cell_type": "markdown", "id": "723c8206", "metadata": {"papermill": {"duration": 0.01805, "end_time": "2022-05-12T12:19:10.583298", "exception": false, "start_time": "2022-05-12T12:19:10.565248", "status": "completed"}, "tags": []}, "source": ["For stabilization purposes, we apply a $\\tanh$ activation function on the scaling output.\n", "This prevents sudden large output values for the scaling that can destabilize training.\n", "To still allow scaling factors smaller or larger than -1 and 1 respectively,\n", "we have a learnable parameter per dimension, called `scaling_factor`.\n", "This scales the tanh to different limits.\n", "Below, we visualize the effect of the scaling factor on the output activation of the scaling terms:"]}, {"cell_type": "code", "execution_count": 13, "id": "c1e1275c", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:10.617526Z", "iopub.status.busy": "2022-05-12T12:19:10.617035Z", "iopub.status.idle": "2022-05-12T12:19:11.394850Z", "shell.execute_reply": "2022-05-12T12:19:11.394156Z"}, "papermill": {"duration": 0.797476, "end_time": "2022-05-12T12:19:11.397055", "exception": false, "start_time": "2022-05-12T12:19:10.599579", "status": "completed"}, "tags": []}, "outputs": [{"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"710.613911pt\" height=\"215.984063pt\" viewBox=\"0 0 710.613911 215.984063\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:10.931761</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 215.984063 \n", "L 710.613911 215.984063 \n", "L 710.613911 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g id=\"patch_2\">\n", "    <path d=\"M 32.916406 188.638125 \n", "L 209.126933 188.638125 \n", "L 209.126933 22.318125 \n", "L 32.916406 22.318125 \n", "z\n", "\" style=\"fill: #eaeaf2\"/>\n", "   </g>\n", "   <g id=\"matplotlib.axis_1\">\n", "    <g id=\"xtick_1\">\n", "     <g id=\"line2d_1\">\n", "      <path d=\"M 40.925976 188.638125 \n", "L 40.925976 22.318125 \n", "\" clip-path=\"url(#pece1999bd8)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_1\">\n", "      <!-- \u22125.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(27.570429 206.496406)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-2212\" d=\"M 678 2272 \n", "L 4684 2272 \n", "L 4684 1741 \n", "L 678 1741 \n", "L 678 2272 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-35\" d=\"M 691 4666 \n", "L 3169 4666 \n", "L 3169 4134 \n", "L 1269 4134 \n", "L 1269 2991 \n", "Q 1406 3038 1543 3061 \n", "Q 1681 3084 1819 3084 \n", "Q 2600 3084 3056 2656 \n", "Q 3513 2228 3513 1497 \n", "Q 3513 744 3044 326 \n", "Q 2575 -91 1722 -91 \n", "Q 1428 -91 1123 -41 \n", "Q 819 9 494 109 \n", "L 494 744 \n", "Q 775 591 1075 516 \n", "Q 1375 441 1709 441 \n", "Q 2250 441 2565 725 \n", "Q 2881 1009 2881 1497 \n", "Q 2881 1984 2565 2268 \n", "Q 2250 2553 1709 2553 \n", "Q 1456 2553 1204 2497 \n", "Q 953 2441 691 2322 \n", "L 691 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-2e\" d=\"M 684 794 \n", "L 1344 794 \n", "L 1344 0 \n", "L 684 0 \n", "L 684 794 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-30\" d=\"M 2034 4250 \n", "Q 1547 4250 1301 3770 \n", "Q 1056 3291 1056 2328 \n", "Q 1056 1369 1301 889 \n", "Q 1547 409 2034 409 \n", "Q 2525 409 2770 889 \n", "Q 3016 1369 3016 2328 \n", "Q 3016 3291 2770 3770 \n", "Q 2525 4250 2034 4250 \n", "z\n", "M 2034 4750 \n", "Q 2819 4750 3233 4129 \n", "Q 3647 3509 3647 2328 \n", "Q 3647 1150 3233 529 \n", "Q 2819 -91 2034 -91 \n", "Q 1250 -91 836 529 \n", "Q 422 1150 422 2328 \n", "Q 422 3509 836 4129 \n", "Q 1250 4750 2034 4750 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"83.789062\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"147.412109\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"179.199219\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_2\">\n", "     <g id=\"line2d_2\">\n", "      <path d=\"M 81.013911 188.638125 \n", "L 81.013911 22.318125 \n", "\" clip-path=\"url(#pece1999bd8)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_2\">\n", "      <!-- \u22122.5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(67.658364 206.496406)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-32\" d=\"M 1228 531 \n", "L 3431 531 \n", "L 3431 0 \n", "L 469 0 \n", "L 469 531 \n", "Q 828 903 1448 1529 \n", "Q 2069 2156 2228 2338 \n", "Q 2531 2678 2651 2914 \n", "Q 2772 3150 2772 3378 \n", "Q 2772 3750 2511 3984 \n", "Q 2250 4219 1831 4219 \n", "Q 1534 4219 1204 4116 \n", "Q 875 4013 500 3803 \n", "L 500 4441 \n", "Q 881 4594 1212 4672 \n", "Q 1544 4750 1819 4750 \n", "Q 2544 4750 2975 4387 \n", "Q 3406 4025 3406 3419 \n", "Q 3406 3131 3298 2873 \n", "Q 3191 2616 2906 2266 \n", "Q 2828 2175 2409 1742 \n", "Q 1991 1309 1228 531 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-32\" x=\"83.789062\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"147.412109\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"179.199219\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_3\">\n", "     <g id=\"line2d_3\">\n", "      <path d=\"M 121.101847 188.638125 \n", "L 121.101847 22.318125 \n", "\" clip-path=\"url(#pece1999bd8)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_3\">\n", "      <!-- 0.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(112.355128 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_4\">\n", "     <g id=\"line2d_4\">\n", "      <path d=\"M 161.189783 188.638125 \n", "L 161.189783 22.318125 \n", "\" clip-path=\"url(#pece1999bd8)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_4\">\n", "      <!-- 2.5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(152.443064 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-32\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_5\">\n", "     <g id=\"line2d_5\">\n", "      <path d=\"M 201.277719 188.638125 \n", "L 201.277719 22.318125 \n", "\" clip-path=\"url(#pece1999bd8)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_5\">\n", "      <!-- 5.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(192.531 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-35\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"matplotlib.axis_2\">\n", "    <g id=\"ytick_1\">\n", "     <g id=\"line2d_6\">\n", "      <path d=\"M 32.916406 188.638125 \n", "L 209.126933 188.638125 \n", "\" clip-path=\"url(#pece1999bd8)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_6\">\n", "      <!-- \u22123 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(7.2 192.817266)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-33\" d=\"M 2597 2516 \n", "Q 3050 2419 3304 2112 \n", "Q 3559 1806 3559 1356 \n", "Q 3559 666 3084 287 \n", "Q 2609 -91 1734 -91 \n", "Q 1441 -91 1130 -33 \n", "Q 819 25 488 141 \n", "L 488 750 \n", "Q 750 597 1062 519 \n", "Q 1375 441 1716 441 \n", "Q 2309 441 2620 675 \n", "Q 2931 909 2931 1356 \n", "Q 2931 1769 2642 2001 \n", "Q 2353 2234 1838 2234 \n", "L 1294 2234 \n", "L 1294 2753 \n", "L 1863 2753 \n", "Q 2328 2753 2575 2939 \n", "Q 2822 3125 2822 3475 \n", "Q 2822 3834 2567 4026 \n", "Q 2313 4219 1838 4219 \n", "Q 1578 4219 1281 4162 \n", "Q 984 4106 628 3988 \n", "L 628 4550 \n", "Q 988 4650 1302 4700 \n", "Q 1616 4750 1894 4750 \n", "Q 2613 4750 3031 4423 \n", "Q 3450 4097 3450 3541 \n", "Q 3450 3153 3228 2886 \n", "Q 3006 2619 2597 2516 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-33\" x=\"83.789062\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_2\">\n", "     <g id=\"line2d_7\">\n", "      <path d=\"M 32.916406 160.918125 \n", "L 209.126933 160.918125 \n", "\" clip-path=\"url(#pece1999bd8)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_7\">\n", "      <!-- \u22122 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(7.2 165.097266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-32\" x=\"83.789062\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_3\">\n", "     <g id=\"line2d_8\">\n", "      <path d=\"M 32.916406 133.198125 \n", "L 209.126933 133.198125 \n", "\" clip-path=\"url(#pece1999bd8)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_8\">\n", "      <!-- \u22121 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(7.2 137.377266)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-31\" d=\"M 794 531 \n", "L 1825 531 \n", "L 1825 4091 \n", "L 703 3866 \n", "L 703 4441 \n", "L 1819 4666 \n", "L 2450 4666 \n", "L 2450 531 \n", "L 3481 531 \n", "L 3481 0 \n", "L 794 0 \n", "L 794 531 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"83.789062\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_4\">\n", "     <g id=\"line2d_9\">\n", "      <path d=\"M 32.916406 105.478125 \n", "L 209.126933 105.478125 \n", "\" clip-path=\"url(#pece1999bd8)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_9\">\n", "      <!-- 0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(16.417656 109.657266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_5\">\n", "     <g id=\"line2d_10\">\n", "      <path d=\"M 32.916406 77.758125 \n", "L 209.126933 77.758125 \n", "\" clip-path=\"url(#pece1999bd8)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_10\">\n", "      <!-- 1 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(16.417656 81.937266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-31\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_6\">\n", "     <g id=\"line2d_11\">\n", "      <path d=\"M 32.916406 50.038125 \n", "L 209.126933 50.038125 \n", "\" clip-path=\"url(#pece1999bd8)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_11\">\n", "      <!-- 2 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(16.417656 54.217266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-32\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_7\">\n", "     <g id=\"line2d_12\">\n", "      <path d=\"M 32.916406 22.318125 \n", "L 209.126933 22.318125 \n", "\" clip-path=\"url(#pece1999bd8)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_12\">\n", "      <!-- 3 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(16.417656 26.497266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-33\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"line2d_13\">\n", "    <path d=\"M 40.925976 119.338125 \n", "L 98.973307 119.227526 \n", "L 103.623509 118.988377 \n", "L 106.509838 118.629025 \n", "L 108.594412 118.165853 \n", "L 110.35828 117.559503 \n", "L 111.961797 116.765905 \n", "L 113.404964 115.793803 \n", "L 114.848129 114.52464 \n", "L 116.291295 112.921632 \n", "L 117.73446 110.979581 \n", "L 119.49833 108.213747 \n", "L 124.950289 99.293189 \n", "L 126.553806 97.279666 \n", "L 127.996972 95.827994 \n", "L 129.440137 94.696596 \n", "L 130.883304 93.840518 \n", "L 132.486821 93.148284 \n", "L 134.25069 92.623355 \n", "L 136.495615 92.201391 \n", "L 139.381946 91.905131 \n", "L 143.551091 91.720252 \n", "L 151.408326 91.632556 \n", "L 186.84606 91.618127 \n", "L 201.117363 91.618125 \n", "L 201.117363 91.618125 \n", "\" clip-path=\"url(#pece1999bd8)\" style=\"fill: none; stroke: #4c72b0; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"patch_3\">\n", "    <path d=\"M 32.916406 188.638125 \n", "L 32.916406 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_4\">\n", "    <path d=\"M 209.126933 188.638125 \n", "L 209.126933 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_5\">\n", "    <path d=\"M 32.916406 188.638125 \n", "L 209.126933 188.638125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_6\">\n", "    <path d=\"M 32.916406 22.318125 \n", "L 209.126933 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"text_13\">\n", "    <!-- Scaling factor: 0.5 -->\n", "    <g style=\"fill: #262626\" transform=\"translate(66.441357 16.318125)scale(0.12 -0.12)\">\n", "     <defs>\n", "      <path id=\"DejaVuSans-53\" d=\"M 3425 4513 \n", "L 3425 3897 \n", "Q 3066 4069 2747 4153 \n", "Q 2428 4238 2131 4238 \n", "Q 1616 4238 1336 4038 \n", "Q 1056 3838 1056 3469 \n", "Q 1056 3159 1242 3001 \n", "Q 1428 2844 1947 2747 \n", "L 2328 2669 \n", "Q 3034 2534 3370 2195 \n", "Q 3706 1856 3706 1288 \n", "Q 3706 609 3251 259 \n", "Q 2797 -91 1919 -91 \n", "Q 1588 -91 1214 -16 \n", "Q 841 59 441 206 \n", "L 441 856 \n", "Q 825 641 1194 531 \n", "Q 1563 422 1919 422 \n", "Q 2459 422 2753 634 \n", "Q 3047 847 3047 1241 \n", "Q 3047 1584 2836 1778 \n", "Q 2625 1972 2144 2069 \n", "L 1759 2144 \n", "Q 1053 2284 737 2584 \n", "Q 422 2884 422 3419 \n", "Q 422 4038 858 4394 \n", "Q 1294 4750 2059 4750 \n", "Q 2388 4750 2728 4690 \n", "Q 3069 4631 3425 4513 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-63\" d=\"M 3122 3366 \n", "L 3122 2828 \n", "Q 2878 2963 2633 3030 \n", "Q 2388 3097 2138 3097 \n", "Q 1578 3097 1268 2742 \n", "Q 959 2388 959 1747 \n", "Q 959 1106 1268 751 \n", "Q 1578 397 2138 397 \n", "Q 2388 397 2633 464 \n", "Q 2878 531 3122 666 \n", "L 3122 134 \n", "Q 2881 22 2623 -34 \n", "Q 2366 -91 2075 -91 \n", "Q 1284 -91 818 406 \n", "Q 353 903 353 1747 \n", "Q 353 2603 823 3093 \n", "Q 1294 3584 2113 3584 \n", "Q 2378 3584 2631 3529 \n", "Q 2884 3475 3122 3366 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-61\" d=\"M 2194 1759 \n", "Q 1497 1759 1228 1600 \n", "Q 959 1441 959 1056 \n", "Q 959 750 1161 570 \n", "Q 1363 391 1709 391 \n", "Q 2188 391 2477 730 \n", "Q 2766 1069 2766 1631 \n", "L 2766 1759 \n", "L 2194 1759 \n", "z\n", "M 3341 1997 \n", "L 3341 0 \n", "L 2766 0 \n", "L 2766 531 \n", "Q 2569 213 2275 61 \n", "Q 1981 -91 1556 -91 \n", "Q 1019 -91 701 211 \n", "Q 384 513 384 1019 \n", "Q 384 1609 779 1909 \n", "Q 1175 2209 1959 2209 \n", "L 2766 2209 \n", "L 2766 2266 \n", "Q 2766 2663 2505 2880 \n", "Q 2244 3097 1772 3097 \n", "Q 1472 3097 1187 3025 \n", "Q 903 2953 641 2809 \n", "L 641 3341 \n", "Q 956 3463 1253 3523 \n", "Q 1550 3584 1831 3584 \n", "Q 2591 3584 2966 3190 \n", "Q 3341 2797 3341 1997 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6c\" d=\"M 603 4863 \n", "L 1178 4863 \n", "L 1178 0 \n", "L 603 0 \n", "L 603 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-69\" d=\"M 603 3500 \n", "L 1178 3500 \n", "L 1178 0 \n", "L 603 0 \n", "L 603 3500 \n", "z\n", "M 603 4863 \n", "L 1178 4863 \n", "L 1178 4134 \n", "L 603 4134 \n", "L 603 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6e\" d=\"M 3513 2113 \n", "L 3513 0 \n", "L 2938 0 \n", "L 2938 2094 \n", "Q 2938 2591 2744 2837 \n", "Q 2550 3084 2163 3084 \n", "Q 1697 3084 1428 2787 \n", "Q 1159 2491 1159 1978 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 3500 \n", "L 1159 3500 \n", "L 1159 2956 \n", "Q 1366 3272 1645 3428 \n", "Q 1925 3584 2291 3584 \n", "Q 2894 3584 3203 3211 \n", "Q 3513 2838 3513 2113 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-67\" d=\"M 2906 1791 \n", "Q 2906 2416 2648 2759 \n", "Q 2391 3103 1925 3103 \n", "Q 1463 3103 1205 2759 \n", "Q 947 2416 947 1791 \n", "Q 947 1169 1205 825 \n", "Q 1463 481 1925 481 \n", "Q 2391 481 2648 825 \n", "Q 2906 1169 2906 1791 \n", "z\n", "M 3481 434 \n", "Q 3481 -459 3084 -895 \n", "Q 2688 -1331 1869 -1331 \n", "Q 1566 -1331 1297 -1286 \n", "Q 1028 -1241 775 -1147 \n", "L 775 -588 \n", "Q 1028 -725 1275 -790 \n", "Q 1522 -856 1778 -856 \n", "Q 2344 -856 2625 -561 \n", "Q 2906 -266 2906 331 \n", "L 2906 616 \n", "Q 2728 306 2450 153 \n", "Q 2172 0 1784 0 \n", "Q 1141 0 747 490 \n", "Q 353 981 353 1791 \n", "Q 353 2603 747 3093 \n", "Q 1141 3584 1784 3584 \n", "Q 2172 3584 2450 3431 \n", "Q 2728 3278 2906 2969 \n", "L 2906 3500 \n", "L 3481 3500 \n", "L 3481 434 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-20\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-66\" d=\"M 2375 4863 \n", "L 2375 4384 \n", "L 1825 4384 \n", "Q 1516 4384 1395 4259 \n", "Q 1275 4134 1275 3809 \n", "L 1275 3500 \n", "L 2222 3500 \n", "L 2222 3053 \n", "L 1275 3053 \n", "L 1275 0 \n", "L 697 0 \n", "L 697 3053 \n", "L 147 3053 \n", "L 147 3500 \n", "L 697 3500 \n", "L 697 3744 \n", "Q 697 4328 969 4595 \n", "Q 1241 4863 1831 4863 \n", "L 2375 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-74\" d=\"M 1172 4494 \n", "L 1172 3500 \n", "L 2356 3500 \n", "L 2356 3053 \n", "L 1172 3053 \n", "L 1172 1153 \n", "Q 1172 725 1289 603 \n", "Q 1406 481 1766 481 \n", "L 2356 481 \n", "L 2356 0 \n", "L 1766 0 \n", "Q 1100 0 847 248 \n", "Q 594 497 594 1153 \n", "L 594 3053 \n", "L 172 3053 \n", "L 172 3500 \n", "L 594 3500 \n", "L 594 4494 \n", "L 1172 4494 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6f\" d=\"M 1959 3097 \n", "Q 1497 3097 1228 2736 \n", "Q 959 2375 959 1747 \n", "Q 959 1119 1226 758 \n", "Q 1494 397 1959 397 \n", "Q 2419 397 2687 759 \n", "Q 2956 1122 2956 1747 \n", "Q 2956 2369 2687 2733 \n", "Q 2419 3097 1959 3097 \n", "z\n", "M 1959 3584 \n", "Q 2709 3584 3137 3096 \n", "Q 3566 2609 3566 1747 \n", "Q 3566 888 3137 398 \n", "Q 2709 -91 1959 -91 \n", "Q 1206 -91 779 398 \n", "Q 353 888 353 1747 \n", "Q 353 2609 779 3096 \n", "Q 1206 3584 1959 3584 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-72\" d=\"M 2631 2963 \n", "Q 2534 3019 2420 3045 \n", "Q 2306 3072 2169 3072 \n", "Q 1681 3072 1420 2755 \n", "Q 1159 2438 1159 1844 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 3500 \n", "L 1159 3500 \n", "L 1159 2956 \n", "Q 1341 3275 1631 3429 \n", "Q 1922 3584 2338 3584 \n", "Q 2397 3584 2469 3576 \n", "Q 2541 3569 2628 3553 \n", "L 2631 2963 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-3a\" d=\"M 750 794 \n", "L 1409 794 \n", "L 1409 0 \n", "L 750 0 \n", "L 750 794 \n", "z\n", "M 750 3309 \n", "L 1409 3309 \n", "L 1409 2516 \n", "L 750 2516 \n", "L 750 3309 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "     </defs>\n", "     <use xlink:href=\"#DejaVuSans-53\"/>\n", "     <use xlink:href=\"#DejaVuSans-63\" x=\"63.476562\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"118.457031\"/>\n", "     <use xlink:href=\"#DejaVuSans-6c\" x=\"179.736328\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"207.519531\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"235.302734\"/>\n", "     <use xlink:href=\"#DejaVuSans-67\" x=\"298.681641\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"362.158203\"/>\n", "     <use xlink:href=\"#DejaVuSans-66\" x=\"393.945312\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"429.150391\"/>\n", "     <use xlink:href=\"#DejaVuSans-63\" x=\"490.429688\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"545.410156\"/>\n", "     <use xlink:href=\"#DejaVuSans-6f\" x=\"584.619141\"/>\n", "     <use xlink:href=\"#DejaVuSans-72\" x=\"645.800781\"/>\n", "     <use xlink:href=\"#DejaVuSans-3a\" x=\"685.164062\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"718.855469\"/>\n", "     <use xlink:href=\"#DejaVuSans-30\" x=\"750.642578\"/>\n", "     <use xlink:href=\"#DejaVuSans-2e\" x=\"814.265625\"/>\n", "     <use xlink:href=\"#DejaVuSans-35\" x=\"846.052734\"/>\n", "    </g>\n", "   </g>\n", "  </g>\n", "  <g id=\"axes_2\">\n", "   <g id=\"patch_7\">\n", "    <path d=\"M 279.611143 188.638125 \n", "L 455.821669 188.638125 \n", "L 455.821669 22.318125 \n", "L 279.611143 22.318125 \n", "z\n", "\" style=\"fill: #eaeaf2\"/>\n", "   </g>\n", "   <g id=\"matplotlib.axis_3\">\n", "    <g id=\"xtick_6\">\n", "     <g id=\"line2d_14\">\n", "      <path d=\"M 287.620712 188.638125 \n", "L 287.620712 22.318125 \n", "\" clip-path=\"url(#p40f90781a3)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_14\">\n", "      <!-- \u22125.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(274.265166 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"83.789062\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"147.412109\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"179.199219\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_7\">\n", "     <g id=\"line2d_15\">\n", "      <path d=\"M 327.708648 188.638125 \n", "L 327.708648 22.318125 \n", "\" clip-path=\"url(#p40f90781a3)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_15\">\n", "      <!-- \u22122.5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(314.353101 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-32\" x=\"83.789062\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"147.412109\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"179.199219\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_8\">\n", "     <g id=\"line2d_16\">\n", "      <path d=\"M 367.796584 188.638125 \n", "L 367.796584 22.318125 \n", "\" clip-path=\"url(#p40f90781a3)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_16\">\n", "      <!-- 0.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(359.049865 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_9\">\n", "     <g id=\"line2d_17\">\n", "      <path d=\"M 407.88452 188.638125 \n", "L 407.88452 22.318125 \n", "\" clip-path=\"url(#p40f90781a3)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_17\">\n", "      <!-- 2.5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(399.137801 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-32\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_10\">\n", "     <g id=\"line2d_18\">\n", "      <path d=\"M 447.972455 188.638125 \n", "L 447.972455 22.318125 \n", "\" clip-path=\"url(#p40f90781a3)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_18\">\n", "      <!-- 5.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(439.225737 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-35\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"matplotlib.axis_4\">\n", "    <g id=\"ytick_8\">\n", "     <g id=\"line2d_19\">\n", "      <path d=\"M 279.611143 188.638125 \n", "L 455.821669 188.638125 \n", "\" clip-path=\"url(#p40f90781a3)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_19\">\n", "      <!-- \u22123 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(253.894737 192.817266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-33\" x=\"83.789062\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_9\">\n", "     <g id=\"line2d_20\">\n", "      <path d=\"M 279.611143 160.918125 \n", "L 455.821669 160.918125 \n", "\" clip-path=\"url(#p40f90781a3)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_20\">\n", "      <!-- \u22122 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(253.894737 165.097266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-32\" x=\"83.789062\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_10\">\n", "     <g id=\"line2d_21\">\n", "      <path d=\"M 279.611143 133.198125 \n", "L 455.821669 133.198125 \n", "\" clip-path=\"url(#p40f90781a3)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_21\">\n", "      <!-- \u22121 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(253.894737 137.377266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"83.789062\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_11\">\n", "     <g id=\"line2d_22\">\n", "      <path d=\"M 279.611143 105.478125 \n", "L 455.821669 105.478125 \n", "\" clip-path=\"url(#p40f90781a3)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_22\">\n", "      <!-- 0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(263.112393 109.657266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_12\">\n", "     <g id=\"line2d_23\">\n", "      <path d=\"M 279.611143 77.758125 \n", "L 455.821669 77.758125 \n", "\" clip-path=\"url(#p40f90781a3)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_23\">\n", "      <!-- 1 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(263.112393 81.937266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-31\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_13\">\n", "     <g id=\"line2d_24\">\n", "      <path d=\"M 279.611143 50.038125 \n", "L 455.821669 50.038125 \n", "\" clip-path=\"url(#p40f90781a3)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_24\">\n", "      <!-- 2 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(263.112393 54.217266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-32\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_14\">\n", "     <g id=\"line2d_25\">\n", "      <path d=\"M 279.611143 22.318125 \n", "L 455.821669 22.318125 \n", "\" clip-path=\"url(#p40f90781a3)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_25\">\n", "      <!-- 3 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(263.112393 26.497266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-33\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"line2d_26\">\n", "    <path d=\"M 287.620712 133.195609 \n", "L 318.728952 133.07651 \n", "L 327.548297 132.834373 \n", "L 333.160606 132.470457 \n", "L 337.329751 131.985028 \n", "L 340.697138 131.372686 \n", "L 343.583471 130.618475 \n", "L 345.988747 129.771741 \n", "L 348.233671 128.75334 \n", "L 350.318245 127.567586 \n", "L 352.242464 126.232209 \n", "L 354.166685 124.634571 \n", "L 356.090906 122.749497 \n", "L 358.015127 120.561328 \n", "L 360.0997 117.847998 \n", "L 362.344625 114.555798 \n", "L 365.070604 110.145648 \n", "L 374.371006 94.709662 \n", "L 376.615929 91.603706 \n", "L 378.700503 89.081207 \n", "L 380.785075 86.917084 \n", "L 382.709296 85.226062 \n", "L 384.793868 83.699523 \n", "L 386.87844 82.454479 \n", "L 389.123366 81.382533 \n", "L 391.528642 80.489433 \n", "L 394.254621 79.730192 \n", "L 397.301303 79.122084 \n", "L 400.989394 78.627052 \n", "L 405.479242 78.257822 \n", "L 411.57261 77.992943 \n", "L 421.033362 77.830491 \n", "L 440.75663 77.764314 \n", "L 447.8121 77.760693 \n", "L 447.8121 77.760693 \n", "\" clip-path=\"url(#p40f90781a3)\" style=\"fill: none; stroke: #4c72b0; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"patch_8\">\n", "    <path d=\"M 279.611143 188.638125 \n", "L 279.611143 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_9\">\n", "    <path d=\"M 455.821669 188.638125 \n", "L 455.821669 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_10\">\n", "    <path d=\"M 279.611143 188.638125 \n", "L 455.821669 188.638125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_11\">\n", "    <path d=\"M 279.611143 22.318125 \n", "L 455.821669 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"text_26\">\n", "    <!-- Scaling factor: 1 -->\n", "    <g style=\"fill: #262626\" transform=\"translate(318.860469 16.318125)scale(0.12 -0.12)\">\n", "     <use xlink:href=\"#DejaVuSans-53\"/>\n", "     <use xlink:href=\"#DejaVuSans-63\" x=\"63.476562\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"118.457031\"/>\n", "     <use xlink:href=\"#DejaVuSans-6c\" x=\"179.736328\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"207.519531\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"235.302734\"/>\n", "     <use xlink:href=\"#DejaVuSans-67\" x=\"298.681641\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"362.158203\"/>\n", "     <use xlink:href=\"#DejaVuSans-66\" x=\"393.945312\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"429.150391\"/>\n", "     <use xlink:href=\"#DejaVuSans-63\" x=\"490.429688\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"545.410156\"/>\n", "     <use xlink:href=\"#DejaVuSans-6f\" x=\"584.619141\"/>\n", "     <use xlink:href=\"#DejaVuSans-72\" x=\"645.800781\"/>\n", "     <use xlink:href=\"#DejaVuSans-3a\" x=\"685.164062\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"718.855469\"/>\n", "     <use xlink:href=\"#DejaVuSans-31\" x=\"750.642578\"/>\n", "    </g>\n", "   </g>\n", "  </g>\n", "  <g id=\"axes_3\">\n", "   <g id=\"patch_12\">\n", "    <path d=\"M 526.30588 188.638125 \n", "L 702.516406 188.638125 \n", "L 702.516406 22.318125 \n", "L 526.30588 22.318125 \n", "z\n", "\" style=\"fill: #eaeaf2\"/>\n", "   </g>\n", "   <g id=\"matplotlib.axis_5\">\n", "    <g id=\"xtick_11\">\n", "     <g id=\"line2d_27\">\n", "      <path d=\"M 534.315449 188.638125 \n", "L 534.315449 22.318125 \n", "\" clip-path=\"url(#p915da914b1)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_27\">\n", "      <!-- \u22125.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(520.959902 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"83.789062\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"147.412109\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"179.199219\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_12\">\n", "     <g id=\"line2d_28\">\n", "      <path d=\"M 574.403385 188.638125 \n", "L 574.403385 22.318125 \n", "\" clip-path=\"url(#p915da914b1)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_28\">\n", "      <!-- \u22122.5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(561.047838 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-32\" x=\"83.789062\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"147.412109\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"179.199219\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_13\">\n", "     <g id=\"line2d_29\">\n", "      <path d=\"M 614.491321 188.638125 \n", "L 614.491321 22.318125 \n", "\" clip-path=\"url(#p915da914b1)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_29\">\n", "      <!-- 0.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(605.744602 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_14\">\n", "     <g id=\"line2d_30\">\n", "      <path d=\"M 654.579257 188.638125 \n", "L 654.579257 22.318125 \n", "\" clip-path=\"url(#p915da914b1)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_30\">\n", "      <!-- 2.5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(645.832538 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-32\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_15\">\n", "     <g id=\"line2d_31\">\n", "      <path d=\"M 694.667192 188.638125 \n", "L 694.667192 22.318125 \n", "\" clip-path=\"url(#p915da914b1)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_31\">\n", "      <!-- 5.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(685.920474 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-35\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"matplotlib.axis_6\">\n", "    <g id=\"ytick_15\">\n", "     <g id=\"line2d_32\">\n", "      <path d=\"M 526.30588 188.638125 \n", "L 702.516406 188.638125 \n", "\" clip-path=\"url(#p915da914b1)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_32\">\n", "      <!-- \u22123 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(500.589474 192.817266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-33\" x=\"83.789062\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_16\">\n", "     <g id=\"line2d_33\">\n", "      <path d=\"M 526.30588 160.918125 \n", "L 702.516406 160.918125 \n", "\" clip-path=\"url(#p915da914b1)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_33\">\n", "      <!-- \u22122 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(500.589474 165.097266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-32\" x=\"83.789062\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_17\">\n", "     <g id=\"line2d_34\">\n", "      <path d=\"M 526.30588 133.198125 \n", "L 702.516406 133.198125 \n", "\" clip-path=\"url(#p915da914b1)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_34\">\n", "      <!-- \u22121 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(500.589474 137.377266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"83.789062\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_18\">\n", "     <g id=\"line2d_35\">\n", "      <path d=\"M 526.30588 105.478125 \n", "L 702.516406 105.478125 \n", "\" clip-path=\"url(#p915da914b1)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_35\">\n", "      <!-- 0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(509.80713 109.657266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_19\">\n", "     <g id=\"line2d_36\">\n", "      <path d=\"M 526.30588 77.758125 \n", "L 702.516406 77.758125 \n", "\" clip-path=\"url(#p915da914b1)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_36\">\n", "      <!-- 1 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(509.80713 81.937266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-31\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_20\">\n", "     <g id=\"line2d_37\">\n", "      <path d=\"M 526.30588 50.038125 \n", "L 702.516406 50.038125 \n", "\" clip-path=\"url(#p915da914b1)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_37\">\n", "      <!-- 2 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(509.80713 54.217266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-32\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_21\">\n", "     <g id=\"line2d_38\">\n", "      <path d=\"M 526.30588 22.318125 \n", "L 702.516406 22.318125 \n", "\" clip-path=\"url(#p915da914b1)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_38\">\n", "      <!-- 3 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(509.80713 26.497266)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-33\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"line2d_39\">\n", "    <path d=\"M 534.315449 160.176021 \n", "L 542.493392 159.687784 \n", "L 548.907463 159.092637 \n", "L 554.359417 158.370393 \n", "L 559.009617 157.539385 \n", "L 563.178762 156.575428 \n", "L 567.027204 155.45551 \n", "L 570.554943 154.192821 \n", "L 573.761979 152.812668 \n", "L 576.808663 151.264271 \n", "L 579.694991 149.555474 \n", "L 582.581324 147.584043 \n", "L 585.307303 145.457699 \n", "L 588.033284 143.054366 \n", "L 590.759263 140.357932 \n", "L 593.485241 137.357585 \n", "L 596.371574 133.845542 \n", "L 599.418257 129.771895 \n", "L 602.625292 125.103426 \n", "L 606.313382 119.316667 \n", "L 610.803231 111.825767 \n", "L 623.952073 89.581798 \n", "L 627.640164 83.9412 \n", "L 630.847198 79.424364 \n", "L 633.89388 75.50738 \n", "L 636.780215 72.148454 \n", "L 639.666543 69.133481 \n", "L 642.392524 66.593944 \n", "L 645.118503 64.339521 \n", "L 648.004834 62.242973 \n", "L 650.891166 60.42073 \n", "L 653.93785 58.765528 \n", "L 657.144882 57.286941 \n", "L 660.512269 55.987576 \n", "L 664.040008 54.863805 \n", "L 667.88845 53.869721 \n", "L 672.217947 52.9872 \n", "L 677.028502 52.238017 \n", "L 682.640812 51.5975 \n", "L 689.215231 51.077923 \n", "L 694.506837 50.787638 \n", "L 694.506837 50.787638 \n", "\" clip-path=\"url(#p915da914b1)\" style=\"fill: none; stroke: #4c72b0; stroke-width: 1.5; stroke-linecap: round\"/>\n", "   </g>\n", "   <g id=\"patch_13\">\n", "    <path d=\"M 526.30588 188.638125 \n", "L 526.30588 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_14\">\n", "    <path d=\"M 702.516406 188.638125 \n", "L 702.516406 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_15\">\n", "    <path d=\"M 526.30588 188.638125 \n", "L 702.516406 188.638125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_16\">\n", "    <path d=\"M 526.30588 22.318125 \n", "L 702.516406 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"text_39\">\n", "    <!-- Scaling factor: 2 -->\n", "    <g style=\"fill: #262626\" transform=\"translate(565.555206 16.318125)scale(0.12 -0.12)\">\n", "     <use xlink:href=\"#DejaVuSans-53\"/>\n", "     <use xlink:href=\"#DejaVuSans-63\" x=\"63.476562\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"118.457031\"/>\n", "     <use xlink:href=\"#DejaVuSans-6c\" x=\"179.736328\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"207.519531\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"235.302734\"/>\n", "     <use xlink:href=\"#DejaVuSans-67\" x=\"298.681641\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"362.158203\"/>\n", "     <use xlink:href=\"#DejaVuSans-66\" x=\"393.945312\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"429.150391\"/>\n", "     <use xlink:href=\"#DejaVuSans-63\" x=\"490.429688\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"545.410156\"/>\n", "     <use xlink:href=\"#DejaVuSans-6f\" x=\"584.619141\"/>\n", "     <use xlink:href=\"#DejaVuSans-72\" x=\"645.800781\"/>\n", "     <use xlink:href=\"#DejaVuSans-3a\" x=\"685.164062\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"718.855469\"/>\n", "     <use xlink:href=\"#DejaVuSans-32\" x=\"750.642578\"/>\n", "    </g>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"pece1999bd8\">\n", "   <rect x=\"32.916406\" y=\"22.318125\" width=\"176.210526\" height=\"166.32\"/>\n", "  </clipPath>\n", "  <clipPath id=\"p40f90781a3\">\n", "   <rect x=\"279.611143\" y=\"22.318125\" width=\"176.210526\" height=\"166.32\"/>\n", "  </clipPath>\n", "  <clipPath id=\"p915da914b1\">\n", "   <rect x=\"526.30588\" y=\"22.318125\" width=\"176.210526\" height=\"166.32\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 1200x300 with 3 Axes>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["with torch.no_grad():\n", "    x = torch.arange(-5, 5, 0.01)\n", "    scaling_factors = [0.5, 1, 2]\n", "    sns.set()\n", "    fig, ax = plt.subplots(1, 3, figsize=(12, 3))\n", "    for i, scale in enumerate(scaling_factors):\n", "        y = torch.tanh(x / scale) * scale\n", "        ax[i].plot(x.numpy(), y.numpy())\n", "        ax[i].set_title(\"Scaling factor: \" + str(scale))\n", "        ax[i].set_ylim(-3, 3)\n", "    plt.subplots_adjust(wspace=0.4)\n", "    sns.reset_orig()\n", "    plt.show()"]}, {"cell_type": "markdown", "id": "fbe411de", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.017149, "end_time": "2022-05-12T12:19:11.433168", "exception": false, "start_time": "2022-05-12T12:19:11.416019", "status": "completed"}, "tags": []}, "source": ["Coupling layers generalize to any masking technique we could think of.\n", "However, the most common approach for images is to split the input $z$ in half, using a checkerboard mask or channel mask.\n", "A checkerboard mask splits the variables across the height and width dimensions and assigns each other pixel to $z_{j+1:d}$.\n", "Thereby, the mask is shared across channels.\n", "In contrast, the channel mask assigns half of the channels to $z_{j+1:d}$, and the other half to $z_{1:j+1}$.\n", "Note that when we apply multiple coupling layers, we invert the masking for each other layer so that each variable is transformed a similar amount of times.\n", "\n", "Let's implement a function that creates a checkerboard mask and a channel mask for us:"]}, {"cell_type": "code", "execution_count": 14, "id": "d0e8846d", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:11.470735Z", "iopub.status.busy": "2022-05-12T12:19:11.470295Z", "iopub.status.idle": "2022-05-12T12:19:11.476167Z", "shell.execute_reply": "2022-05-12T12:19:11.475559Z"}, "papermill": {"duration": 0.026902, "end_time": "2022-05-12T12:19:11.477658", "exception": false, "start_time": "2022-05-12T12:19:11.450756", "status": "completed"}, "tags": []}, "outputs": [], "source": ["def create_checkerboard_mask(h, w, invert=False):\n", "    x, y = torch.arange(h, dtype=torch.int32), torch.arange(w, dtype=torch.int32)\n", "    xx, yy = torch.meshgrid(x, y)\n", "    mask = torch.fmod(xx + yy, 2)\n", "    mask = mask.to(torch.float32).view(1, 1, h, w)\n", "    if invert:\n", "        mask = 1 - mask\n", "    return mask\n", "\n", "\n", "def create_channel_mask(c_in, invert=False):\n", "    mask = torch.cat([torch.ones(c_in // 2, dtype=torch.float32), torch.zeros(c_in - c_in // 2, dtype=torch.float32)])\n", "    mask = mask.view(1, c_in, 1, 1)\n", "    if invert:\n", "        mask = 1 - mask\n", "    return mask"]}, {"cell_type": "markdown", "id": "082a68c8", "metadata": {"papermill": {"duration": 0.017254, "end_time": "2022-05-12T12:19:11.512191", "exception": false, "start_time": "2022-05-12T12:19:11.494937", "status": "completed"}, "tags": []}, "source": ["We can also visualize the corresponding masks for an image of size $8\\times 8\\times 2$ (2 channels):"]}, {"cell_type": "code", "execution_count": 15, "id": "55168db2", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:11.548462Z", "iopub.status.busy": "2022-05-12T12:19:11.547815Z", "iopub.status.idle": "2022-05-12T12:19:11.669378Z", "shell.execute_reply": "2022-05-12T12:19:11.668705Z"}, "papermill": {"duration": 0.141383, "end_time": "2022-05-12T12:19:11.670995", "exception": false, "start_time": "2022-05-12T12:19:11.529612", "status": "completed"}, "tags": []}, "outputs": [{"data": {"application/pdf": "JVBERi0xLjQKJazcIKu6CjEgMCBvYmoKPDwgL1BhZ2VzIDIgMCBSIC9UeXBlIC9DYXRhbG9nID4+CmVuZG9iago4IDAgb2JqCjw8IC9FeHRHU3RhdGUgNCAwIFIgL0ZvbnQgMyAwIFIgL1BhdHRlcm4gNSAwIFIKL1Byb2NTZXQgWyAvUERGIC9UZXh0IC9JbWFnZUIgL0ltYWdlQyAvSW1hZ2VJIF0gL1NoYWRpbmcgNiAwIFIKL1hPYmplY3QgNyAwIFIgPj4KZW5kb2JqCjExIDAgb2JqCjw8IC9Bbm5vdHMgMTAgMCBSIC9Db250ZW50cyA5IDAgUiAvTWVkaWFCb3ggWyAwIDAgMTY2Ljg2IDExMi42ODUgXQovUGFyZW50IDIgMCBSIC9SZXNvdXJjZXMgOCAwIFIgL1R5cGUgL1BhZ2UgPj4KZW5kb2JqCjkgMCBvYmoKPDwgL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAxMiAwIFIgPj4Kc3RyZWFtCnicVU/LTsMwELzvV8wxPeB47XqTHGkLEdwKkTggDiV106aEKolEf59tJB61NNKM7ZmdTVfx61DHp3KB5TOlf6oeidEqGli0ijMYpaIhq6ojFjG5KP34oczOSB70wv7je6Id9ciMm8DBmbkg94YFQ8QLPpHeauyo2a3irJElrpv0k03muIxWa3C/eXWH9IGxOmFNa/T63lwFqaZe17C40VZwXmNs4bOAQowX9dOiQnrPYIdqNy1XbekVyXIf6+MMPhgO4gKSOLyfNsMMLKZwmc395SDZotuM+vEN1SPdVaQ96BvN3kySCmVuZHN0cmVhbQplbmRvYmoKMTIgMCBvYmoKMjMwCmVuZG9iagoxMCAwIG9iagpbIF0KZW5kb2JqCjE4IDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMjM1ID4+CnN0cmVhbQp4nDVRSW4AMQi75xX+QKWwJ++Zquqh/f+1hlEvAwPY2CTvwUYkPsSQ7ihXfMrqNMvwO1nkxc9K4eS9iAqkKsIKaQfPclYzDJ4bmQKXM/FZZj6ZFjsWUE3EcXbkNINBiGlcR8vpMNM86Am5PhhxY6dZrmJI691Svb7X8p8qykfW3Sy3TtnUSt2iZ+xJXHZeT21pXxh1FDcFkQ4fO7wH+SLmLC46kW72mymHlaQhOC2AH4mhVM8OrxEmfmYkeMqeTu+jNLz2QdP1vXtBR24mZCq3UEYqnqw0xoyh+o1oJqnv/4Ge9b2+/gBDTVS5CmVuZHN0cmVhbQplbmRvYmoKMTkgMCBvYmoKPDwgL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAzMDcgPj4Kc3RyZWFtCnicPZJLbgMxDEP3PoUuEMD62Z7zpCi6mN5/2ycl6Yoc2RZFapa6TFlTHpA0k4R/6fBwsZ3yO2zPZmbgWqKXieWU59AVYu6ifNnMRl1ZJ8XqhGY6t+hRORcHNk2qn6sspd0ueA7XJp5b9hE/vNCgHtQ1Lgk3dFejZSk0Y6r7f9J7/Iwy4GpMXWxSq3sfPF5EVejoB0eJImOXF+fjQQnpSsJoWoiVd0UDQe7ytMp7Ce7b3mrIsgepmM47KWaw63RSLm4XhyEeyPKo8OWj2GtCz/iwKyX0SNiGM3In7mjG5tTI4pD+3o0ES4+uaCHz4K9u1i5gvFM6RWJkTnKsaYtVTvdQFNO5w70MEPVsRUMpc5HV6l/DzgtrlmwWeEr6BR6j3SZLDlbZ26hO76082dD3H1rXdB8KZW5kc3RyZWFtCmVuZG9iagoyMCAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDI0NCA+PgpzdHJlYW0KeJxFkU1yBSEIhPeeoi/wquRXPc+kUllM7r8NzbwkK1qF5gPTAhNH8BJD7ImVEx8yfC/oMny3MjvwOtmZcE+4blzDZcMzYVvgOyrLO15Dd7ZSP52hqu8aOd4uUjV0ZWSfeqGaC8yQiK4RWXQrl3VA05TuUuEabFuCFPVKrCedoDToEcrwd5RrfHUTT6+x5FTNIVrNrRMairBseEHUySQRtQ2LJ5ZzIVH5qhurOi5gkyXi9IDcoJVmfHpSSREwg3ysyWjMAjbQk7tnF8aaSx5Fjlc0mLA7STXwgPfitr73NnGP8xf4hXff/ysOfdcCPn8AS/5dBgplbmRzdHJlYW0KZW5kb2JqCjIxIDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMjMyID4+CnN0cmVhbQp4nDVRSW7EMAy7+xX8wADW7rwnxaCH9v/XUsoUCEAltrglYmMjAi8x+DmI3PiSNaMmfmdyV/wsT4VHwq3gSRSBl+FedoLLG8ZlPw4zH7yXVs6kxpMMyEU2PTwRMtglEDowuwZ12Gbaib4h4bMjUs1GltPXEvTSKgTKU7bf6YISbav6c/usC2372hNOdnvqSeUTiOeWrMBl4xWTxVgGPVG5SzF9kOpsoSehvCifg2w+aohElyhn4InBwSjQDuy57WfiVSFoXd2nbWOoRkrH078NTU2SCPlECWe2NO4W/n/Pvb7X+w9OIVQRCmVuZHN0cmVhbQplbmRvYmoKMjIgMCBvYmoKPDwgL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAyMzEgPj4Kc3RyZWFtCnicNU85kgQhDMt5hT4wVRjbQL+np7Y22Pl/upKZTpDwIcnTEx2ZeJkjI7Bmx9taZCBm4FNMxb/2tA8TqvfgHiKUiwthhpFw1qzjbp6OF/92lc9YB+82+IpZXhDYwkzWVxZnLtsFY2mcxDnJboxdE7GNda2nU1hHMKEMhHS2w5Qgc1Sk9MmOMuboOJEnnovv9tssdjl+DusLNo0hFef4KnqCNoOi7HnvAhpyQf9d3fgeRbvoJSAbCRbWUWLunOWEX712dB61KBJzQppBLhMhzekqphCaUKyzo6BSUXCpPqforJ9/5V9cLQplbmRzdHJlYW0KZW5kb2JqCjIzIDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMjQ5ID4+CnN0cmVhbQp4nD1QO45EIQzrOYUv8CTyI3AeRqstZu/frgOaKVBMfrYzJNARgUcMMZSv4yWtoK6Bv4tC8W7i64PCIKtDUiDOeg+IdOymNpETOh2cMz9hN2OOwEUxBpzpdKY9ByY5+8IKhHMbZexWSCeJqiKO6jOOKZ4qe594FiztyDZbJ5I95CDhUlKJyaWflMo/bcqUCjpm0QQsErngZBNNOMu7SVKMGZQy6h6mdiJ9rDzIozroZE3OrCOZ2dNP25n4HHC3X9pkTpXHdB7M+Jy0zoM5Fbr344k2B02N2ujs9xNpKi9Sux1anX51EpXdGOcYEpdnfxnfZP/5B/6HWiIKZW5kc3RyZWFtCmVuZG9iagoyNCAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDE2NCA+PgpzdHJlYW0KeJxFkMdxBTEMQ++qAiUwgAr1rMfzD+v+r4b000F6GEIMYk/CsFxXcWF0w4+3LTMNf0cZ7sb6MmO81VggJ+gDDJGJq9Gk+nbFGar05NVirqOiXC86IhLMkuOrQCN8OrLHk7a2M/10Xh/sIe8T/yoq525hAS6q7kD5Uh/x1I/ZUeqaoY8qK2seatpXhF0RSts+LqcyTt29A1rhvZWrPdrvPx52OvIKZW5kc3RyZWFtCmVuZG9iagoyNSAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDgzID4+CnN0cmVhbQp4nD3MORKAMAgF0J5T/COEyCL3cRyLeP9WMNEGHqt6oCE4g7rBreFgyrp0E+9T49XGnBIJqHhKTZa6C3rUtL7Uvmjgu+vmS9WJP83PF50Pux0Z3QplbmRzdHJlYW0KZW5kb2JqCjI2IDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMjU4ID4+CnN0cmVhbQp4nEWRS3IEIAhE956CI4D85DyTSmUxuf82Dc5kNnaXqP2ESiOmEiznFHkwfcnyzWS26Xc5VjsbBRRFKJjJVeixAqs7U8SZa4lq62Nl5LjTOwbFG85dOalkcaOMdVR1KnBMz5X1Ud35dlmUfUcOZQrYrHMcbODKbcMYJ0abre4O94kgTydTR8XtINnwByeNfZWrK3CdbPbRSzAOBP1CE5jki0DrDIHGzVP05BLs4+N254Fgb3kRSNkQyJEhGB2Cdp1c/+LW+b3/cYY7z7UZrhzv4neY1nbHX2KSFXMBi9wpqOdrLlrXGTrekzPH5Kb7hs65YJe7g0zv+T/Wz/r+Ax4pZvoKZW5kc3RyZWFtCmVuZG9iagoyNyAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDIxOCA+PgpzdHJlYW0KeJw9ULmNBDEMy12FGljAeu2pZxaLS6b/9Ej59iLRFkVSKjWZkikvdZQlWVPeOnyWxA55huVuZDYlKkUvk7Al99AK8X2J5hT33dWWs0M0l2g5fgszKqobHdNLNppwKhO6oNzDM/oNbXQDVocesVsg0KRg17YgcscPGAzBmROLIgxKTQb/rnKPn16LGz7D8UMUkZIO5jX/WP3ycw2vU48nkW5vvuJenKkOAxEckpq8I11YsS4SEWk1QU3PwFotgLu3Xv4btCO6DED2icRxmlKOob9rcKXPL+UnU9gKZW5kc3RyZWFtCmVuZG9iagoyOCAwIG9iago8PCAvRmlsdGVyIC9GbGF0ZURlY29kZSAvTGVuZ3RoIDE2MCA+PgpzdHJlYW0KeJxFkDkSAzEIBHO9gidIXIL3rMu1wfr/qQfWR6LpAjQcuhZNynoUaD7psUahutBr6CxKkkTBFpIdUKdjiDsoSExIY5JIth6DI5pYs12YmVQqs1LhtGnFwr/ZWtXIRI1wjfyJ6QZU/E/qXJTwTYOvkjH6GFS8O4OMSfheRdxaMe3+RDCxGfYJb0UmBYSJsanZvs9ghsz3Ctc4x/MNTII36wplbmRzdHJlYW0KZW5kb2JqCjI5IDAgb2JqCjw8IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlIC9MZW5ndGggMzM0ID4+CnN0cmVhbQp4nC1SS3LFIAzbcwpdoDP4B+Q86XS6eL3/tpKTRUYOYPQx5YaJSnxZILej1sS3jcxAheGvq8yFz0jbyDqIy5CLuJIthXtELOQxxDzEgu+r8R4e+azMybMHxi/Zdw8r9tSEZSHjxRnaYRXHYRXkWLB1Iap7eFOkw6kk2OOL/z7Fcy0ELXxG0IBf5J+vjuD5khZp95ht0656sEw7qqSwHGxPc14mX1pnuToezwfJ9q7YEVK7AhSFuTPOc+Eo01ZGtBZ2NkhqXGxvjv1YStCFblxGiiOQn6kiPKCkycwmCuKPnB5yKgNh6pqudHIbVXGnnsw1m4u3M0lm675IsZnCeV04s/4MU2a1eSfPcqLUqQjvsWdL0NA5rp69lllodJsTvKSEz8ZOT06+VzPrITkVCaliWlfBaRSZYgnbEl9TUVOaehn++/Lu8Tt+/gEsc3xzCmVuZHN0cmVhbQplbmRvYmoKMzAgMCBvYmoKPDwgL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0xlbmd0aCAxOCA+PgpzdHJlYW0KeJwzNrRQMIDDFEOuNAAd5gNSCmVuZHN0cmVhbQplbmRvYmoKMTYgMCBvYmoKPDwgL0Jhc2VGb250IC9CTVFRRFYrRGVqYVZ1U2FucyAvQ2hhclByb2NzIDE3IDAgUgovRW5jb2RpbmcgPDwKL0RpZmZlcmVuY2VzIFsgMzIgL3NwYWNlIDY3IC9DIDk3IC9hIC9iIC9jIC9kIC9lIDEwNCAvaCAxMDcgL2sgMTA5IC9tIDExMSAvbyAxMTQgL3IgL3MKXQovVHlwZSAvRW5jb2RpbmcgPj4KL0ZpcnN0Q2hhciAwIC9Gb250QkJveCBbIC0xMDIxIC00NjMgMTc5NCAxMjMzIF0gL0ZvbnREZXNjcmlwdG9yIDE1IDAgUgovRm9udE1hdHJpeCBbIDAuMDAxIDAgMCAwLjAwMSAwIDAgXSAvTGFzdENoYXIgMjU1IC9OYW1lIC9CTVFRRFYrRGVqYVZ1U2FucwovU3VidHlwZSAvVHlwZTMgL1R5cGUgL0ZvbnQgL1dpZHRocyAxNCAwIFIgPj4KZW5kb2JqCjE1IDAgb2JqCjw8IC9Bc2NlbnQgOTI5IC9DYXBIZWlnaHQgMCAvRGVzY2VudCAtMjM2IC9GbGFncyAzMgovRm9udEJCb3ggWyAtMTAyMSAtNDYzIDE3OTQgMTIzMyBdIC9Gb250TmFtZSAvQk1RUURWK0RlamFWdVNhbnMKL0l0YWxpY0FuZ2xlIDAgL01heFdpZHRoIDEzNDIgL1N0ZW1WIDAgL1R5cGUgL0ZvbnREZXNjcmlwdG9yIC9YSGVpZ2h0IDAgPj4KZW5kb2JqCjE0IDAgb2JqClsgNjAwIDYwMCA2MDAgNjAwIDYwMCA2MDAgNjAwIDYwMCA2MDAgNjAwIDYwMCA2MDAgNjAwIDYwMCA2MDAgNjAwIDYwMCA2MDAKNjAwIDYwMCA2MDAgNjAwIDYwMCA2MDAgNjAwIDYwMCA2MDAgNjAwIDYwMCA2MDAgNjAwIDYwMCAzMTggNDAxIDQ2MCA4MzggNjM2Cjk1MCA3ODAgMjc1IDM5MCAzOTAgNTAwIDgzOCAzMTggMzYxIDMxOCAzMzcgNjM2IDYzNiA2MzYgNjM2IDYzNiA2MzYgNjM2IDYzNgo2MzYgNjM2IDMzNyAzMzcgODM4IDgzOCA4MzggNTMxIDEwMDAgNjg0IDY4NiA2OTggNzcwIDYzMiA1NzUgNzc1IDc1MiAyOTUKMjk1IDY1NiA1NTcgODYzIDc0OCA3ODcgNjAzIDc4NyA2OTUgNjM1IDYxMSA3MzIgNjg0IDk4OSA2ODUgNjExIDY4NSAzOTAgMzM3CjM5MCA4MzggNTAwIDUwMCA2MTMgNjM1IDU1MCA2MzUgNjE1IDM1MiA2MzUgNjM0IDI3OCAyNzggNTc5IDI3OCA5NzQgNjM0IDYxMgo2MzUgNjM1IDQxMSA1MjEgMzkyIDYzNCA1OTIgODE4IDU5MiA1OTIgNTI1IDYzNiAzMzcgNjM2IDgzOCA2MDAgNjM2IDYwMCAzMTgKMzUyIDUxOCAxMDAwIDUwMCA1MDAgNTAwIDEzNDIgNjM1IDQwMCAxMDcwIDYwMCA2ODUgNjAwIDYwMCAzMTggMzE4IDUxOCA1MTgKNTkwIDUwMCAxMDAwIDUwMCAxMDAwIDUyMSA0MDAgMTAyMyA2MDAgNTI1IDYxMSAzMTggNDAxIDYzNiA2MzYgNjM2IDYzNiAzMzcKNTAwIDUwMCAxMDAwIDQ3MSA2MTIgODM4IDM2MSAxMDAwIDUwMCA1MDAgODM4IDQwMSA0MDEgNTAwIDYzNiA2MzYgMzE4IDUwMAo0MDEgNDcxIDYxMiA5NjkgOTY5IDk2OSA1MzEgNjg0IDY4NCA2ODQgNjg0IDY4NCA2ODQgOTc0IDY5OCA2MzIgNjMyIDYzMiA2MzIKMjk1IDI5NSAyOTUgMjk1IDc3NSA3NDggNzg3IDc4NyA3ODcgNzg3IDc4NyA4MzggNzg3IDczMiA3MzIgNzMyIDczMiA2MTEgNjA1CjYzMCA2MTMgNjEzIDYxMyA2MTMgNjEzIDYxMyA5ODIgNTUwIDYxNSA2MTUgNjE1IDYxNSAyNzggMjc4IDI3OCAyNzggNjEyIDYzNAo2MTIgNjEyIDYxMiA2MTIgNjEyIDgzOCA2MTIgNjM0IDYzNCA2MzQgNjM0IDU5MiA2MzUgNTkyIF0KZW5kb2JqCjE3IDAgb2JqCjw8IC9DIDE4IDAgUiAvYSAxOSAwIFIgL2IgMjAgMCBSIC9jIDIxIDAgUiAvZCAyMiAwIFIgL2UgMjMgMCBSIC9oIDI0IDAgUgovayAyNSAwIFIgL20gMjYgMCBSIC9vIDI3IDAgUiAvciAyOCAwIFIgL3MgMjkgMCBSIC9zcGFjZSAzMCAwIFIgPj4KZW5kb2JqCjMgMCBvYmoKPDwgL0YxIDE2IDAgUiA+PgplbmRvYmoKNCAwIG9iago8PCAvQTEgPDwgL0NBIDEgL1R5cGUgL0V4dEdTdGF0ZSAvY2EgMSA+PiA+PgplbmRvYmoKNSAwIG9iago8PCA+PgplbmRvYmoKNiAwIG9iago8PCA+PgplbmRvYmoKNyAwIG9iago8PCAvSTEgMTMgMCBSID4+CmVuZG9iagoxMyAwIG9iago8PCAvQml0c1BlckNvbXBvbmVudCA4IC9Db2xvclNwYWNlIFsvSW5kZXhlZCAvRGV2aWNlUkdCIDIgKP///39/fwAAACldCi9EZWNvZGVQYXJtcyA8PCAvQ29sb3JzIDEgL0NvbHVtbnMgMjEyIC9QcmVkaWN0b3IgMTAgPj4KL0ZpbHRlciAvRmxhdGVEZWNvZGUgL0hlaWdodCAxMTYgL0xlbmd0aCAzMSAwIFIgL1N1YnR5cGUgL0ltYWdlCi9UeXBlIC9YT2JqZWN0IC9XaWR0aCAyMTIgPj4Kc3RyZWFtCnic7dAxTkAxDETBD/c/NJ1pXCxSENnPvNKS5Uyejxf2/PUDfiOolqBagmoJqiWolqBagmoJqqV/g/qcnumHo+/ZdjTbzEbbASgoKCgoKKgLUCdOzWhDZZvh90FBQUFBQUHdiTpxakYbKtsMvw8KCgoKCuoFqBOnZrShss3w+6CgoKCgoKDuRJ04NaMNlW2G3wcFBQUFBQV1J+rEqRltqGwz/D4oKCgoKKgXoE6cmtGGyjbD74OCgoKCgoK6E3Xi1Iw2VLYZfh8UFBQUFBTUnaj6oFqCagmqJaiWoFqCagmqJaiWoFr6AjRXYBEKZW5kc3RyZWFtCmVuZG9iagozMSAwIG9iagoyMzUKZW5kb2JqCjIgMCBvYmoKPDwgL0NvdW50IDEgL0tpZHMgWyAxMSAwIFIgXSAvVHlwZSAvUGFnZXMgPj4KZW5kb2JqCjMyIDAgb2JqCjw8IC9DcmVhdGlvbkRhdGUgKEQ6MjAyMjA1MTIxNDE5MTErMDInMDAnKQovQ3JlYXRvciAoTWF0cGxvdGxpYiB2My41LjIsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcpCi9Qcm9kdWNlciAoTWF0cGxvdGxpYiBwZGYgYmFja2VuZCB2My41LjIpID4+CmVuZG9iagp4cmVmCjAgMzMKMDAwMDAwMDAwMCA2NTUzNSBmIAowMDAwMDAwMDE2IDAwMDAwIG4gCjAwMDAwMDY4MzIgMDAwMDAgbiAKMDAwMDAwNjE1NyAwMDAwMCBuIAowMDAwMDA2MTg5IDAwMDAwIG4gCjAwMDAwMDYyNDkgMDAwMDAgbiAKMDAwMDAwNjI3MCAwMDAwMCBuIAowMDAwMDA2MjkxIDAwMDAwIG4gCjAwMDAwMDAwNjUgMDAwMDAgbiAKMDAwMDAwMDMzNyAwMDAwMCBuIAowMDAwMDAwNjYyIDAwMDAwIG4gCjAwMDAwMDAyMDggMDAwMDAgbiAKMDAwMDAwMDY0MiAwMDAwMCBuIAowMDAwMDA2MzIzIDAwMDAwIG4gCjAwMDAwMDQ5NDggMDAwMDAgbiAKMDAwMDAwNDc0MSAwMDAwMCBuIAowMDAwMDA0MzYyIDAwMDAwIG4gCjAwMDAwMDYwMDEgMDAwMDAgbiAKMDAwMDAwMDY4MiAwMDAwMCBuIAowMDAwMDAwOTkwIDAwMDAwIG4gCjAwMDAwMDEzNzAgMDAwMDAgbiAKMDAwMDAwMTY4NyAwMDAwMCBuIAowMDAwMDAxOTkyIDAwMDAwIG4gCjAwMDAwMDIyOTYgMDAwMDAgbiAKMDAwMDAwMjYxOCAwMDAwMCBuIAowMDAwMDAyODU1IDAwMDAwIG4gCjAwMDAwMDMwMTAgMDAwMDAgbiAKMDAwMDAwMzM0MSAwMDAwMCBuIAowMDAwMDAzNjMyIDAwMDAwIG4gCjAwMDAwMDM4NjUgMDAwMDAgbiAKMDAwMDAwNDI3MiAwMDAwMCBuIAowMDAwMDA2ODEyIDAwMDAwIG4gCjAwMDAwMDY4OTIgMDAwMDAgbiAKdHJhaWxlcgo8PCAvSW5mbyAzMiAwIFIgL1Jvb3QgMSAwIFIgL1NpemUgMzMgPj4Kc3RhcnR4cmVmCjcwNDkKJSVFT0YK\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"166.86pt\" height=\"112.678125pt\" viewBox=\"0 0 166.86 112.678125\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:11.575759</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 112.678125 \n", "L 166.86 112.678125 \n", "L 166.86 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g clip-path=\"url(#pc9ec646cb7)\">\n", "    <image xlink:href=\"data:image/png;base64,\n", "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\" id=\"imagec48e239167\" transform=\"scale(1 -1)translate(0 -83.52)\" x=\"7.2\" y=\"-21.958125\" width=\"152.64\" height=\"83.52\"/>\n", "   </g>\n", "   <g id=\"text_1\">\n", "    <!-- Checkerboard mask -->\n", "    <g transform=\"translate(23.675625 16.318125)scale(0.12 -0.12)\">\n", "     <defs>\n", "      <path id=\"DejaVuSans-43\" d=\"M 4122 4306 \n", "L 4122 3641 \n", "Q 3803 3938 3442 4084 \n", "Q 3081 4231 2675 4231 \n", "Q 1875 4231 1450 3742 \n", "Q 1025 3253 1025 2328 \n", "Q 1025 1406 1450 917 \n", "Q 1875 428 2675 428 \n", "Q 3081 428 3442 575 \n", "Q 3803 722 4122 1019 \n", "L 4122 359 \n", "Q 3791 134 3420 21 \n", "Q 3050 -91 2638 -91 \n", "Q 1578 -91 968 557 \n", "Q 359 1206 359 2328 \n", "Q 359 3453 968 4101 \n", "Q 1578 4750 2638 4750 \n", "Q 3056 4750 3426 4639 \n", "Q 3797 4528 4122 4306 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-68\" d=\"M 3513 2113 \n", "L 3513 0 \n", "L 2938 0 \n", "L 2938 2094 \n", "Q 2938 2591 2744 2837 \n", "Q 2550 3084 2163 3084 \n", "Q 1697 3084 1428 2787 \n", "Q 1159 2491 1159 1978 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 4863 \n", "L 1159 4863 \n", "L 1159 2956 \n", "Q 1366 3272 1645 3428 \n", "Q 1925 3584 2291 3584 \n", "Q 2894 3584 3203 3211 \n", "Q 3513 2838 3513 2113 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-65\" d=\"M 3597 1894 \n", "L 3597 1613 \n", "L 953 1613 \n", "Q 991 1019 1311 708 \n", "Q 1631 397 2203 397 \n", "Q 2534 397 2845 478 \n", "Q 3156 559 3463 722 \n", "L 3463 178 \n", "Q 3153 47 2828 -22 \n", "Q 2503 -91 2169 -91 \n", "Q 1331 -91 842 396 \n", "Q 353 884 353 1716 \n", "Q 353 2575 817 3079 \n", "Q 1281 3584 2069 3584 \n", "Q 2775 3584 3186 3129 \n", "Q 3597 2675 3597 1894 \n", "z\n", "M 3022 2063 \n", "Q 3016 2534 2758 2815 \n", "Q 2500 3097 2075 3097 \n", "Q 1594 3097 1305 2825 \n", "Q 1016 2553 972 2059 \n", "L 3022 2063 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-63\" d=\"M 3122 3366 \n", "L 3122 2828 \n", "Q 2878 2963 2633 3030 \n", "Q 2388 3097 2138 3097 \n", "Q 1578 3097 1268 2742 \n", "Q 959 2388 959 1747 \n", "Q 959 1106 1268 751 \n", "Q 1578 397 2138 397 \n", "Q 2388 397 2633 464 \n", "Q 2878 531 3122 666 \n", "L 3122 134 \n", "Q 2881 22 2623 -34 \n", "Q 2366 -91 2075 -91 \n", "Q 1284 -91 818 406 \n", "Q 353 903 353 1747 \n", "Q 353 2603 823 3093 \n", "Q 1294 3584 2113 3584 \n", "Q 2378 3584 2631 3529 \n", "Q 2884 3475 3122 3366 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6b\" d=\"M 581 4863 \n", "L 1159 4863 \n", "L 1159 1991 \n", "L 2875 3500 \n", "L 3609 3500 \n", "L 1753 1863 \n", "L 3688 0 \n", "L 2938 0 \n", "L 1159 1709 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-72\" d=\"M 2631 2963 \n", "Q 2534 3019 2420 3045 \n", "Q 2306 3072 2169 3072 \n", "Q 1681 3072 1420 2755 \n", "Q 1159 2438 1159 1844 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 3500 \n", "L 1159 3500 \n", "L 1159 2956 \n", "Q 1341 3275 1631 3429 \n", "Q 1922 3584 2338 3584 \n", "Q 2397 3584 2469 3576 \n", "Q 2541 3569 2628 3553 \n", "L 2631 2963 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-62\" d=\"M 3116 1747 \n", "Q 3116 2381 2855 2742 \n", "Q 2594 3103 2138 3103 \n", "Q 1681 3103 1420 2742 \n", "Q 1159 2381 1159 1747 \n", "Q 1159 1113 1420 752 \n", "Q 1681 391 2138 391 \n", "Q 2594 391 2855 752 \n", "Q 3116 1113 3116 1747 \n", "z\n", "M 1159 2969 \n", "Q 1341 3281 1617 3432 \n", "Q 1894 3584 2278 3584 \n", "Q 2916 3584 3314 3078 \n", "Q 3713 2572 3713 1747 \n", "Q 3713 922 3314 415 \n", "Q 2916 -91 2278 -91 \n", "Q 1894 -91 1617 61 \n", "Q 1341 213 1159 525 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 4863 \n", "L 1159 4863 \n", "L 1159 2969 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6f\" d=\"M 1959 3097 \n", "Q 1497 3097 1228 2736 \n", "Q 959 2375 959 1747 \n", "Q 959 1119 1226 758 \n", "Q 1494 397 1959 397 \n", "Q 2419 397 2687 759 \n", "Q 2956 1122 2956 1747 \n", "Q 2956 2369 2687 2733 \n", "Q 2419 3097 1959 3097 \n", "z\n", "M 1959 3584 \n", "Q 2709 3584 3137 3096 \n", "Q 3566 2609 3566 1747 \n", "Q 3566 888 3137 398 \n", "Q 2709 -91 1959 -91 \n", "Q 1206 -91 779 398 \n", "Q 353 888 353 1747 \n", "Q 353 2609 779 3096 \n", "Q 1206 3584 1959 3584 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-61\" d=\"M 2194 1759 \n", "Q 1497 1759 1228 1600 \n", "Q 959 1441 959 1056 \n", "Q 959 750 1161 570 \n", "Q 1363 391 1709 391 \n", "Q 2188 391 2477 730 \n", "Q 2766 1069 2766 1631 \n", "L 2766 1759 \n", "L 2194 1759 \n", "z\n", "M 3341 1997 \n", "L 3341 0 \n", "L 2766 0 \n", "L 2766 531 \n", "Q 2569 213 2275 61 \n", "Q 1981 -91 1556 -91 \n", "Q 1019 -91 701 211 \n", "Q 384 513 384 1019 \n", "Q 384 1609 779 1909 \n", "Q 1175 2209 1959 2209 \n", "L 2766 2209 \n", "L 2766 2266 \n", "Q 2766 2663 2505 2880 \n", "Q 2244 3097 1772 3097 \n", "Q 1472 3097 1187 3025 \n", "Q 903 2953 641 2809 \n", "L 641 3341 \n", "Q 956 3463 1253 3523 \n", "Q 1550 3584 1831 3584 \n", "Q 2591 3584 2966 3190 \n", "Q 3341 2797 3341 1997 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-64\" d=\"M 2906 2969 \n", "L 2906 4863 \n", "L 3481 4863 \n", "L 3481 0 \n", "L 2906 0 \n", "L 2906 525 \n", "Q 2725 213 2448 61 \n", "Q 2172 -91 1784 -91 \n", "Q 1150 -91 751 415 \n", "Q 353 922 353 1747 \n", "Q 353 2572 751 3078 \n", "Q 1150 3584 1784 3584 \n", "Q 2172 3584 2448 3432 \n", "Q 2725 3281 2906 2969 \n", "z\n", "M 947 1747 \n", "Q 947 1113 1208 752 \n", "Q 1469 391 1925 391 \n", "Q 2381 391 2643 752 \n", "Q 2906 1113 2906 1747 \n", "Q 2906 2381 2643 2742 \n", "Q 2381 3103 1925 3103 \n", "Q 1469 3103 1208 2742 \n", "Q 947 2381 947 1747 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-20\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6d\" d=\"M 3328 2828 \n", "Q 3544 3216 3844 3400 \n", "Q 4144 3584 4550 3584 \n", "Q 5097 3584 5394 3201 \n", "Q 5691 2819 5691 2113 \n", "L 5691 0 \n", "L 5113 0 \n", "L 5113 2094 \n", "Q 5113 2597 4934 2840 \n", "Q 4756 3084 4391 3084 \n", "Q 3944 3084 3684 2787 \n", "Q 3425 2491 3425 1978 \n", "L 3425 0 \n", "L 2847 0 \n", "L 2847 2094 \n", "Q 2847 2600 2669 2842 \n", "Q 2491 3084 2119 3084 \n", "Q 1678 3084 1418 2786 \n", "Q 1159 2488 1159 1978 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 3500 \n", "L 1159 3500 \n", "L 1159 2956 \n", "Q 1356 3278 1631 3431 \n", "Q 1906 3584 2284 3584 \n", "Q 2666 3584 2933 3390 \n", "Q 3200 3197 3328 2828 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-73\" d=\"M 2834 3397 \n", "L 2834 2853 \n", "Q 2591 2978 2328 3040 \n", "Q 2066 3103 1784 3103 \n", "Q 1356 3103 1142 2972 \n", "Q 928 2841 928 2578 \n", "Q 928 2378 1081 2264 \n", "Q 1234 2150 1697 2047 \n", "L 1894 2003 \n", "Q 2506 1872 2764 1633 \n", "Q 3022 1394 3022 966 \n", "Q 3022 478 2636 193 \n", "Q 2250 -91 1575 -91 \n", "Q 1294 -91 989 -36 \n", "Q 684 19 347 128 \n", "L 347 722 \n", "Q 666 556 975 473 \n", "Q 1284 391 1588 391 \n", "Q 1994 391 2212 530 \n", "Q 2431 669 2431 922 \n", "Q 2431 1156 2273 1281 \n", "Q 2116 1406 1581 1522 \n", "L 1381 1569 \n", "Q 847 1681 609 1914 \n", "Q 372 2147 372 2553 \n", "Q 372 3047 722 3315 \n", "Q 1072 3584 1716 3584 \n", "Q 2034 3584 2315 3537 \n", "Q 2597 3491 2834 3397 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "     </defs>\n", "     <use xlink:href=\"#DejaVuSans-43\"/>\n", "     <use xlink:href=\"#DejaVuSans-68\" x=\"69.824219\"/>\n", "     <use xlink:href=\"#DejaVuSans-65\" x=\"133.203125\"/>\n", "     <use xlink:href=\"#DejaVuSans-63\" x=\"194.726562\"/>\n", "     <use xlink:href=\"#DejaVuSans-6b\" x=\"249.707031\"/>\n", "     <use xlink:href=\"#DejaVuSans-65\" x=\"303.992188\"/>\n", "     <use xlink:href=\"#DejaVuSans-72\" x=\"365.515625\"/>\n", "     <use xlink:href=\"#DejaVuSans-62\" x=\"406.628906\"/>\n", "     <use xlink:href=\"#DejaVuSans-6f\" x=\"470.105469\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"531.287109\"/>\n", "     <use xlink:href=\"#DejaVuSans-72\" x=\"592.566406\"/>\n", "     <use xlink:href=\"#DejaVuSans-64\" x=\"631.929688\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"695.40625\"/>\n", "     <use xlink:href=\"#DejaVuSans-6d\" x=\"727.193359\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"824.605469\"/>\n", "     <use xlink:href=\"#DejaVuSans-73\" x=\"885.884766\"/>\n", "     <use xlink:href=\"#DejaVuSans-6b\" x=\"937.984375\"/>\n", "    </g>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"pc9ec646cb7\">\n", "   <rect x=\"7.2\" y=\"22.318125\" width=\"152.46\" height=\"83.16\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 300x150 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}, {"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"166.86pt\" height=\"112.678125pt\" viewBox=\"0 0 166.86 112.678125\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:11.634536</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 112.678125 \n", "L 166.86 112.678125 \n", "L 166.86 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g clip-path=\"url(#p5e686135ef)\">\n", "    <image xlink:href=\"data:image/png;base64,\n", "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\" id=\"image24adf8cc66\" transform=\"scale(1 -1)translate(0 -83.52)\" x=\"7.2\" y=\"-21.958125\" width=\"152.64\" height=\"83.52\"/>\n", "   </g>\n", "   <g id=\"text_1\">\n", "    <!-- Channel mask -->\n", "    <g transform=\"translate(40.769063 16.318125)scale(0.12 -0.12)\">\n", "     <defs>\n", "      <path id=\"DejaVuSans-43\" d=\"M 4122 4306 \n", "L 4122 3641 \n", "Q 3803 3938 3442 4084 \n", "Q 3081 4231 2675 4231 \n", "Q 1875 4231 1450 3742 \n", "Q 1025 3253 1025 2328 \n", "Q 1025 1406 1450 917 \n", "Q 1875 428 2675 428 \n", "Q 3081 428 3442 575 \n", "Q 3803 722 4122 1019 \n", "L 4122 359 \n", "Q 3791 134 3420 21 \n", "Q 3050 -91 2638 -91 \n", "Q 1578 -91 968 557 \n", "Q 359 1206 359 2328 \n", "Q 359 3453 968 4101 \n", "Q 1578 4750 2638 4750 \n", "Q 3056 4750 3426 4639 \n", "Q 3797 4528 4122 4306 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-68\" d=\"M 3513 2113 \n", "L 3513 0 \n", "L 2938 0 \n", "L 2938 2094 \n", "Q 2938 2591 2744 2837 \n", "Q 2550 3084 2163 3084 \n", "Q 1697 3084 1428 2787 \n", "Q 1159 2491 1159 1978 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 4863 \n", "L 1159 4863 \n", "L 1159 2956 \n", "Q 1366 3272 1645 3428 \n", "Q 1925 3584 2291 3584 \n", "Q 2894 3584 3203 3211 \n", "Q 3513 2838 3513 2113 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-61\" d=\"M 2194 1759 \n", "Q 1497 1759 1228 1600 \n", "Q 959 1441 959 1056 \n", "Q 959 750 1161 570 \n", "Q 1363 391 1709 391 \n", "Q 2188 391 2477 730 \n", "Q 2766 1069 2766 1631 \n", "L 2766 1759 \n", "L 2194 1759 \n", "z\n", "M 3341 1997 \n", "L 3341 0 \n", "L 2766 0 \n", "L 2766 531 \n", "Q 2569 213 2275 61 \n", "Q 1981 -91 1556 -91 \n", "Q 1019 -91 701 211 \n", "Q 384 513 384 1019 \n", "Q 384 1609 779 1909 \n", "Q 1175 2209 1959 2209 \n", "L 2766 2209 \n", "L 2766 2266 \n", "Q 2766 2663 2505 2880 \n", "Q 2244 3097 1772 3097 \n", "Q 1472 3097 1187 3025 \n", "Q 903 2953 641 2809 \n", "L 641 3341 \n", "Q 956 3463 1253 3523 \n", "Q 1550 3584 1831 3584 \n", "Q 2591 3584 2966 3190 \n", "Q 3341 2797 3341 1997 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6e\" d=\"M 3513 2113 \n", "L 3513 0 \n", "L 2938 0 \n", "L 2938 2094 \n", "Q 2938 2591 2744 2837 \n", "Q 2550 3084 2163 3084 \n", "Q 1697 3084 1428 2787 \n", "Q 1159 2491 1159 1978 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 3500 \n", "L 1159 3500 \n", "L 1159 2956 \n", "Q 1366 3272 1645 3428 \n", "Q 1925 3584 2291 3584 \n", "Q 2894 3584 3203 3211 \n", "Q 3513 2838 3513 2113 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-65\" d=\"M 3597 1894 \n", "L 3597 1613 \n", "L 953 1613 \n", "Q 991 1019 1311 708 \n", "Q 1631 397 2203 397 \n", "Q 2534 397 2845 478 \n", "Q 3156 559 3463 722 \n", "L 3463 178 \n", "Q 3153 47 2828 -22 \n", "Q 2503 -91 2169 -91 \n", "Q 1331 -91 842 396 \n", "Q 353 884 353 1716 \n", "Q 353 2575 817 3079 \n", "Q 1281 3584 2069 3584 \n", "Q 2775 3584 3186 3129 \n", "Q 3597 2675 3597 1894 \n", "z\n", "M 3022 2063 \n", "Q 3016 2534 2758 2815 \n", "Q 2500 3097 2075 3097 \n", "Q 1594 3097 1305 2825 \n", "Q 1016 2553 972 2059 \n", "L 3022 2063 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6c\" d=\"M 603 4863 \n", "L 1178 4863 \n", "L 1178 0 \n", "L 603 0 \n", "L 603 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-20\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6d\" d=\"M 3328 2828 \n", "Q 3544 3216 3844 3400 \n", "Q 4144 3584 4550 3584 \n", "Q 5097 3584 5394 3201 \n", "Q 5691 2819 5691 2113 \n", "L 5691 0 \n", "L 5113 0 \n", "L 5113 2094 \n", "Q 5113 2597 4934 2840 \n", "Q 4756 3084 4391 3084 \n", "Q 3944 3084 3684 2787 \n", "Q 3425 2491 3425 1978 \n", "L 3425 0 \n", "L 2847 0 \n", "L 2847 2094 \n", "Q 2847 2600 2669 2842 \n", "Q 2491 3084 2119 3084 \n", "Q 1678 3084 1418 2786 \n", "Q 1159 2488 1159 1978 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 3500 \n", "L 1159 3500 \n", "L 1159 2956 \n", "Q 1356 3278 1631 3431 \n", "Q 1906 3584 2284 3584 \n", "Q 2666 3584 2933 3390 \n", "Q 3200 3197 3328 2828 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-73\" d=\"M 2834 3397 \n", "L 2834 2853 \n", "Q 2591 2978 2328 3040 \n", "Q 2066 3103 1784 3103 \n", "Q 1356 3103 1142 2972 \n", "Q 928 2841 928 2578 \n", "Q 928 2378 1081 2264 \n", "Q 1234 2150 1697 2047 \n", "L 1894 2003 \n", "Q 2506 1872 2764 1633 \n", "Q 3022 1394 3022 966 \n", "Q 3022 478 2636 193 \n", "Q 2250 -91 1575 -91 \n", "Q 1294 -91 989 -36 \n", "Q 684 19 347 128 \n", "L 347 722 \n", "Q 666 556 975 473 \n", "Q 1284 391 1588 391 \n", "Q 1994 391 2212 530 \n", "Q 2431 669 2431 922 \n", "Q 2431 1156 2273 1281 \n", "Q 2116 1406 1581 1522 \n", "L 1381 1569 \n", "Q 847 1681 609 1914 \n", "Q 372 2147 372 2553 \n", "Q 372 3047 722 3315 \n", "Q 1072 3584 1716 3584 \n", "Q 2034 3584 2315 3537 \n", "Q 2597 3491 2834 3397 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6b\" d=\"M 581 4863 \n", "L 1159 4863 \n", "L 1159 1991 \n", "L 2875 3500 \n", "L 3609 3500 \n", "L 1753 1863 \n", "L 3688 0 \n", "L 2938 0 \n", "L 1159 1709 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "     </defs>\n", "     <use xlink:href=\"#DejaVuSans-43\"/>\n", "     <use xlink:href=\"#DejaVuSans-68\" x=\"69.824219\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"133.203125\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"194.482422\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"257.861328\"/>\n", "     <use xlink:href=\"#DejaVuSans-65\" x=\"321.240234\"/>\n", "     <use xlink:href=\"#DejaVuSans-6c\" x=\"382.763672\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"410.546875\"/>\n", "     <use xlink:href=\"#DejaVuSans-6d\" x=\"442.333984\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"539.746094\"/>\n", "     <use xlink:href=\"#DejaVuSans-73\" x=\"601.025391\"/>\n", "     <use xlink:href=\"#DejaVuSans-6b\" x=\"653.125\"/>\n", "    </g>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"p5e686135ef\">\n", "   <rect x=\"7.2\" y=\"22.318125\" width=\"152.46\" height=\"83.16\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 300x150 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["checkerboard_mask = create_checkerboard_mask(h=8, w=8).expand(-1, 2, -1, -1)\n", "channel_mask = create_channel_mask(c_in=2).expand(-1, -1, 8, 8)\n", "\n", "show_imgs(checkerboard_mask.transpose(0, 1), \"Checkerboard mask\")\n", "show_imgs(channel_mask.transpose(0, 1), \"Channel mask\")"]}, {"cell_type": "markdown", "id": "01da453b", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.018032, "end_time": "2022-05-12T12:19:11.709109", "exception": false, "start_time": "2022-05-12T12:19:11.691077", "status": "completed"}, "tags": []}, "source": ["As a last aspect of coupling layers, we need to decide for the deep neural network we want to apply in the coupling layers.\n", "The input to the layers is an image, and hence we stick with a CNN.\n", "Because the input to a transformation depends on all transformations before,\n", "it is crucial to ensure a good gradient flow through the CNN back to the input,\n", "which can be optimally achieved by a ResNet-like architecture.\n", "Specifically, we use a Gated ResNet that adds a $\\sigma$-gate to the skip connection,\n", "similarly to the input gate in LSTMs.\n", "The details are not necessarily important here, and the network is\n", "strongly inspired from Flow++ [3] in case you are interested in building\n", "even stronger models."]}, {"cell_type": "code", "execution_count": 16, "id": "9c427445", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:11.748218Z", "iopub.status.busy": "2022-05-12T12:19:11.747617Z", "iopub.status.idle": "2022-05-12T12:19:11.757827Z", "shell.execute_reply": "2022-05-12T12:19:11.757184Z"}, "lines_to_next_cell": 2, "papermill": {"duration": 0.031488, "end_time": "2022-05-12T12:19:11.759391", "exception": false, "start_time": "2022-05-12T12:19:11.727903", "status": "completed"}, "tags": []}, "outputs": [], "source": ["class ConcatELU(nn.Module):\n", "    \"\"\"Activation function that applies ELU in both direction (inverted and plain).\n", "\n", "    Allows non-linearity while providing strong gradients for any input (important for final convolution)\n", "    \"\"\"\n", "\n", "    def forward(self, x):\n", "        return torch.cat([F.elu(x), F.elu(-x)], dim=1)\n", "\n", "\n", "class LayerNormChannels(nn.Module):\n", "    def __init__(self, c_in):\n", "        \"\"\"This module applies layer norm across channels in an image.\n", "\n", "        Has been shown to work well with ResNet connections.\n", "        Args:\n", "            c_in: Number of channels of the input\n", "        \"\"\"\n", "        super().__init__()\n", "        self.layer_norm = nn.LayerNorm(c_in)\n", "\n", "    def forward(self, x):\n", "        x = x.permute(0, 2, 3, 1)\n", "        x = self.layer_norm(x)\n", "        x = x.permute(0, 3, 1, 2)\n", "        return x\n", "\n", "\n", "class GatedConv(nn.Module):\n", "    def __init__(self, c_in, c_hidden):\n", "        \"\"\"\n", "        This module applies a two-layer convolutional ResNet block with input gate\n", "        Args:\n", "            c_in: Number of channels of the input\n", "            c_hidden: Number of hidden dimensions we want to model (usually similar to c_in)\n", "        \"\"\"\n", "        super().__init__()\n", "        self.net = nn.Sequential(\n", "            nn.Conv2d(c_in, c_hidden, kernel_size=3, padding=1),\n", "            ConcatELU(),\n", "            nn.Conv2d(2 * c_hidden, 2 * c_in, kernel_size=1),\n", "        )\n", "\n", "    def forward(self, x):\n", "        out = self.net(x)\n", "        val, gate = out.chunk(2, dim=1)\n", "        return x + val * torch.sigmoid(gate)\n", "\n", "\n", "class GatedConvNet(nn.Module):\n", "    def __init__(self, c_in, c_hidden=32, c_out=-1, num_layers=3):\n", "        \"\"\"Module that summarizes the previous blocks to a full convolutional neural network.\n", "\n", "        Args:\n", "            c_in: Number of input channels\n", "            c_hidden: Number of hidden dimensions to use within the network\n", "            c_out: Number of output channels. If -1, 2 times the input channels are used (affine coupling)\n", "            num_layers: Number of gated ResNet blocks to apply\n", "        \"\"\"\n", "        super().__init__()\n", "        c_out = c_out if c_out > 0 else 2 * c_in\n", "        layers = []\n", "        layers += [nn.Conv2d(c_in, c_hidden, kernel_size=3, padding=1)]\n", "        for layer_index in range(num_layers):\n", "            layers += [GatedConv(c_hidden, c_hidden), LayerNormChannels(c_hidden)]\n", "        layers += [ConcatELU(), nn.Conv2d(2 * c_hidden, c_out, kernel_size=3, padding=1)]\n", "        self.nn = nn.Sequential(*layers)\n", "\n", "        self.nn[-1].weight.data.zero_()\n", "        self.nn[-1].bias.data.zero_()\n", "\n", "    def forward(self, x):\n", "        return self.nn(x)"]}, {"cell_type": "markdown", "id": "cbaa014d", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.018089, "end_time": "2022-05-12T12:19:11.797248", "exception": false, "start_time": "2022-05-12T12:19:11.779159", "status": "completed"}, "tags": []}, "source": ["### Training loop\n", "\n", "Finally, we can add Dequantization, Variational Dequantization and Coupling Layers together to build our full normalizing flow on MNIST images.\n", "We apply 8 coupling layers in the main flow, and 4 for variational dequantization if applied.\n", "We apply a checkerboard mask throughout the network as with a single channel (black-white images),\n", "we cannot apply channel mask.\n", "The overall architecture is visualized below.\n", "\n", "\n", "<center width=\"100%\" style=\"padding: 20px\"><img src=\"https://github.com/PyTorchLightning/lightning-tutorials/raw/main/course_UvA-DL/09-normalizing-flows/vanilla_flow.svg\" width=\"900px\"></center>"]}, {"cell_type": "code", "execution_count": 17, "id": "4baabfa9", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:11.839983Z", "iopub.status.busy": "2022-05-12T12:19:11.839549Z", "iopub.status.idle": "2022-05-12T12:19:11.846018Z", "shell.execute_reply": "2022-05-12T12:19:11.845330Z"}, "lines_to_next_cell": 2, "papermill": {"duration": 0.028068, "end_time": "2022-05-12T12:19:11.847757", "exception": false, "start_time": "2022-05-12T12:19:11.819689", "status": "completed"}, "tags": []}, "outputs": [], "source": ["def create_simple_flow(use_vardeq=True):\n", "    flow_layers = []\n", "    if use_vardeq:\n", "        vardeq_layers = [\n", "            CouplingLayer(\n", "                network=GatedConvNet(c_in=2, c_out=2, c_hidden=16),\n", "                mask=create_checkerboard_mask(h=28, w=28, invert=(i % 2 == 1)),\n", "                c_in=1,\n", "            )\n", "            for i in range(4)\n", "        ]\n", "        flow_layers += [VariationalDequantization(var_flows=vardeq_layers)]\n", "    else:\n", "        flow_layers += [Dequantization()]\n", "\n", "    for i in range(8):\n", "        flow_layers += [\n", "            CouplingLayer(\n", "                network=GatedConvNet(c_in=1, c_hidden=32),\n", "                mask=create_checkerboard_mask(h=28, w=28, invert=(i % 2 == 1)),\n", "                c_in=1,\n", "            )\n", "        ]\n", "\n", "    flow_model = ImageFlow(flow_layers).to(device)\n", "    return flow_model"]}, {"cell_type": "markdown", "id": "6355c7e0", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.019906, "end_time": "2022-05-12T12:19:11.889126", "exception": false, "start_time": "2022-05-12T12:19:11.869220", "status": "completed"}, "tags": []}, "source": ["For implementing the training loop, we use the framework of PyTorch Lightning and reduce the code overhead.\n", "If interested, you can take a look at the generated tensorboard file,\n", "in particularly the graph to see an overview of flow transformations that are applied.\n", "Note that we again provide pre-trained models (see later on in the notebook)\n", "as normalizing flows are particularly expensive to train.\n", "We have also run validation and testing as this can take some time as well with the added importance sampling."]}, {"cell_type": "code", "execution_count": 18, "id": "304541b2", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:11.928154Z", "iopub.status.busy": "2022-05-12T12:19:11.927713Z", "iopub.status.idle": "2022-05-12T12:19:11.935492Z", "shell.execute_reply": "2022-05-12T12:19:11.934876Z"}, "lines_to_next_cell": 2, "papermill": {"duration": 0.029315, "end_time": "2022-05-12T12:19:11.936963", "exception": false, "start_time": "2022-05-12T12:19:11.907648", "status": "completed"}, "tags": []}, "outputs": [], "source": ["def train_flow(flow, model_name=\"MNISTFlow\"):\n", "    # Create a PyTorch Lightning trainer\n", "    trainer = pl.Trainer(\n", "        default_root_dir=os.path.join(CHECKPOINT_PATH, model_name),\n", "        gpus=1 if torch.cuda.is_available() else 0,\n", "        max_epochs=200,\n", "        gradient_clip_val=1.0,\n", "        callbacks=[\n", "            ModelCheckpoint(save_weights_only=True, mode=\"min\", monitor=\"val_bpd\"),\n", "            LearningRateMonitor(\"epoch\"),\n", "        ],\n", "    )\n", "    trainer.logger._log_graph = True\n", "    trainer.logger._default_hp_metric = None  # Optional logging argument that we don't need\n", "\n", "    train_data_loader = data.DataLoader(\n", "        train_set, batch_size=128, shuffle=True, drop_last=True, pin_memory=True, num_workers=8\n", "    )\n", "    result = None\n", "\n", "    # Check whether pretrained model exists. If yes, load it and skip training\n", "    pretrained_filename = os.path.join(CHECKPOINT_PATH, model_name + \".ckpt\")\n", "    if os.path.isfile(pretrained_filename):\n", "        print(\"Found pretrained model, loading...\")\n", "        ckpt = torch.load(pretrained_filename, map_location=device)\n", "        flow.load_state_dict(ckpt[\"state_dict\"])\n", "        result = ckpt.get(\"result\", None)\n", "    else:\n", "        print(\"Start training\", model_name)\n", "        trainer.fit(flow, train_data_loader, val_loader)\n", "\n", "    # Test best model on validation and test set if no result has been found\n", "    # Testing can be expensive due to the importance sampling.\n", "    if result is None:\n", "        val_result = trainer.test(flow, dataloaders=val_loader, verbose=False)\n", "        start_time = time.time()\n", "        test_result = trainer.test(flow, dataloaders=test_loader, verbose=False)\n", "        duration = time.time() - start_time\n", "        result = {\"test\": test_result, \"val\": val_result, \"time\": duration / len(test_loader) / flow.import_samples}\n", "\n", "    return flow, result"]}, {"cell_type": "markdown", "id": "8c41b4f7", "metadata": {"papermill": {"duration": 0.01931, "end_time": "2022-05-12T12:19:11.975188", "exception": false, "start_time": "2022-05-12T12:19:11.955878", "status": "completed"}, "tags": []}, "source": ["## Multi-scale architecture\n", "\n", "<div class=\"center-wrapper\"><div class=\"video-wrapper\"><iframe src=\"https://www.youtube.com/embed/nTyDvn-ADJ4\" title=\"YouTube video player\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture\" allowfullscreen></iframe></div></div>\n", "\n", "One disadvantage of normalizing flows is that they operate on the exact same dimensions as the input.\n", "If the input is high-dimensional, so is the latent space, which requires larger computational cost to learn suitable transformations.\n", "However, particularly in the image domain, many pixels contain less information in the sense\n", "that we could remove them without loosing the semantical information of the image.\n", "\n", "Based on this intuition, deep normalizing flows on images commonly apply a multi-scale architecture [1].\n", "After the first $N$ flow transformations, we split off half of the latent dimensions and directly evaluate them on the prior.\n", "The other half is run through $N$ more flow transformations, and depending on the size of the input,\n", "we split it again in half or stop overall at this position.\n", "The two operations involved in this setup is `Squeeze` and `Split` which\n", "we will review more closely and implement below."]}, {"cell_type": "markdown", "id": "bf4c83d8", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.018063, "end_time": "2022-05-12T12:19:12.011683", "exception": false, "start_time": "2022-05-12T12:19:11.993620", "status": "completed"}, "tags": []}, "source": ["### Squeeze and Split\n", "\n", "When we want to remove half of the pixels in an image, we have the problem of deciding which variables to cut,\n", "and how to rearrange the image.\n", "Thus, the squeezing operation is commonly used before split, which divides the image into subsquares\n", "of shape $2\\times 2\\times C$, and reshapes them into $1\\times 1\\times 4C$ blocks.\n", "Effectively, we reduce the height and width of the image by a factor of 2 while scaling the number of channels by 4.\n", "Afterwards, we can perform the split operation over channels without the need of rearranging the pixels.\n", "The smaller scale also makes the overall architecture more efficient.\n", "Visually, the squeeze operation should transform the input as follows:\n", "\n", "<center><img src=\"https://github.com/PyTorchLightning/lightning-tutorials/raw/main/course_UvA-DL/09-normalizing-flows/Squeeze_operation.svg\" width=\"40%\"/></center>\n", "\n", "The input of $4\\times 4\\times 1$ is scaled to $2\\times 2\\times 4$ following\n", "the idea of grouping the pixels in $2\\times 2\\times 1$ subsquares.\n", "Next, let's try to implement this layer:"]}, {"cell_type": "code", "execution_count": 19, "id": "21bd22ac", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:12.050335Z", "iopub.status.busy": "2022-05-12T12:19:12.049864Z", "iopub.status.idle": "2022-05-12T12:19:12.055719Z", "shell.execute_reply": "2022-05-12T12:19:12.055088Z"}, "papermill": {"duration": 0.027489, "end_time": "2022-05-12T12:19:12.057265", "exception": false, "start_time": "2022-05-12T12:19:12.029776", "status": "completed"}, "tags": []}, "outputs": [], "source": ["class SqueezeFlow(nn.Module):\n", "    def forward(self, z, ldj, reverse=False):\n", "        B, C, H, W = z.shape\n", "        if not reverse:\n", "            # Forward direction: H x W x C => H/2 x W/2 x 4C\n", "            z = z.reshape(B, C, H // 2, 2, W // 2, 2)\n", "            z = z.permute(0, 1, 3, 5, 2, 4)\n", "            z = z.reshape(B, 4 * C, H // 2, W // 2)\n", "        else:\n", "            # Reverse direction: H/2 x W/2 x 4C => H x W x C\n", "            z = z.reshape(B, C // 4, 2, 2, H, W)\n", "            z = z.permute(0, 1, 4, 2, 5, 3)\n", "            z = z.reshape(B, C // 4, H * 2, W * 2)\n", "        return z, ldj"]}, {"cell_type": "markdown", "id": "af45308a", "metadata": {"papermill": {"duration": 0.018839, "end_time": "2022-05-12T12:19:12.095432", "exception": false, "start_time": "2022-05-12T12:19:12.076593", "status": "completed"}, "tags": []}, "source": ["Before moving on, we can verify our implementation by comparing our output with the example figure above:"]}, {"cell_type": "code", "execution_count": 20, "id": "35170bd0", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:12.135426Z", "iopub.status.busy": "2022-05-12T12:19:12.134980Z", "iopub.status.idle": "2022-05-12T12:19:12.141717Z", "shell.execute_reply": "2022-05-12T12:19:12.140991Z"}, "papermill": {"duration": 0.028309, "end_time": "2022-05-12T12:19:12.143258", "exception": false, "start_time": "2022-05-12T12:19:12.114949", "status": "completed"}, "tags": []}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["Image (before)\n", " tensor([[[[ 1,  2,  3,  4],\n", "          [ 5,  6,  7,  8],\n", "          [ 9, 10, 11, 12],\n", "          [13, 14, 15, 16]]]])\n", "\n", "Image (forward)\n", " tensor([[[[ 1,  2,  5,  6],\n", "          [ 3,  4,  7,  8]],\n", "\n", "         [[ 9, 10, 13, 14],\n", "          [11, 12, 15, 16]]]])\n", "\n", "Image (reverse)\n", " tensor([[[[ 1,  2,  3,  4],\n", "          [ 5,  6,  7,  8],\n", "          [ 9, 10, 11, 12],\n", "          [13, 14, 15, 16]]]])\n"]}], "source": ["sq_flow = SqueezeFlow()\n", "rand_img = torch.arange(1, 17).view(1, 1, 4, 4)\n", "print(\"Image (before)\\n\", rand_img)\n", "forward_img, _ = sq_flow(rand_img, ldj=None, reverse=False)\n", "print(\"\\nImage (forward)\\n\", forward_img.permute(0, 2, 3, 1))  # Permute for readability\n", "reconst_img, _ = sq_flow(forward_img, ldj=None, reverse=True)\n", "print(\"\\nImage (reverse)\\n\", reconst_img)"]}, {"cell_type": "markdown", "id": "38bdebf0", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.018912, "end_time": "2022-05-12T12:19:12.181515", "exception": false, "start_time": "2022-05-12T12:19:12.162603", "status": "completed"}, "tags": []}, "source": ["The split operation divides the input into two parts, and evaluates one part directly on the prior.\n", "So that our flow operation fits to the implementation of the previous layers,\n", "we will return the prior probability of the first part as the log determinant jacobian of the layer.\n", "It has the same effect as if we would combine all variable splits at the\n", "end of the flow, and evaluate them together on the prior."]}, {"cell_type": "code", "execution_count": 21, "id": "3c093208", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:12.221775Z", "iopub.status.busy": "2022-05-12T12:19:12.221330Z", "iopub.status.idle": "2022-05-12T12:19:12.227327Z", "shell.execute_reply": "2022-05-12T12:19:12.226649Z"}, "lines_to_next_cell": 2, "papermill": {"duration": 0.028576, "end_time": "2022-05-12T12:19:12.228774", "exception": false, "start_time": "2022-05-12T12:19:12.200198", "status": "completed"}, "tags": []}, "outputs": [], "source": ["class SplitFlow(nn.Module):\n", "    def __init__(self):\n", "        super().__init__()\n", "        self.prior = torch.distributions.normal.Normal(loc=0.0, scale=1.0)\n", "\n", "    def forward(self, z, ldj, reverse=False):\n", "        if not reverse:\n", "            z, z_split = z.chunk(2, dim=1)\n", "            ldj += self.prior.log_prob(z_split).sum(dim=[1, 2, 3])\n", "        else:\n", "            z_split = self.prior.sample(sample_shape=z.shape).to(device)\n", "            z = torch.cat([z, z_split], dim=1)\n", "            ldj -= self.prior.log_prob(z_split).sum(dim=[1, 2, 3])\n", "        return z, ldj"]}, {"cell_type": "markdown", "id": "892077ee", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.019532, "end_time": "2022-05-12T12:19:12.266987", "exception": false, "start_time": "2022-05-12T12:19:12.247455", "status": "completed"}, "tags": []}, "source": ["### Building a multi-scale flow\n", "\n", "After defining the squeeze and split operation, we are finally able to build our own multi-scale flow.\n", "Deep normalizing flows such as Glow and Flow++ [2,3] often apply a split operation directly after squeezing.\n", "However, with shallow flows, we need to be more thoughtful about where to place the split operation as we need at least a minimum amount of transformations on each variable.\n", "Our setup is inspired by the original RealNVP architecture [1] which is shallower than other,\n", "more recent state-of-the-art architectures.\n", "\n", "Hence, for the MNIST dataset, we will apply the first squeeze operation after two coupling layers, but don't apply a split operation yet.\n", "Because we have only used two coupling layers and each the variable has been only transformed once, a split operation would be too early.\n", "We apply two more coupling layers before finally applying a split flow and squeeze again.\n", "The last four coupling layers operate on a scale of $7\\times 7\\times 8$.\n", "The full flow architecture is shown below.\n", "\n", "<center width=\"100%\" style=\"padding: 20px\"><img src=\"https://github.com/PyTorchLightning/lightning-tutorials/raw/main/course_UvA-DL/09-normalizing-flows/multiscale_flow.svg\" width=\"1100px\"></center>\n", "\n", "Note that while the feature maps inside the coupling layers reduce with the height and width of the input,\n", "the increased number of channels is not directly considered.\n", "To counteract this, we increase the hidden dimensions for the coupling layers on the squeezed input.\n", "The dimensions are often scaled by 2 as this approximately increases the computation cost by 4 canceling with the squeezing operation.\n", "However, we will choose the hidden dimensionalities $32, 48, 64$ for the\n", "three scales respectively to keep the number of parameters reasonable\n", "and show the efficiency of multi-scale architectures."]}, {"cell_type": "code", "execution_count": 22, "id": "c89d5d05", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:12.306654Z", "iopub.status.busy": "2022-05-12T12:19:12.306215Z", "iopub.status.idle": "2022-05-12T12:19:12.313447Z", "shell.execute_reply": "2022-05-12T12:19:12.312759Z"}, "lines_to_next_cell": 2, "papermill": {"duration": 0.029036, "end_time": "2022-05-12T12:19:12.314986", "exception": false, "start_time": "2022-05-12T12:19:12.285950", "status": "completed"}, "tags": []}, "outputs": [], "source": ["def create_multiscale_flow():\n", "    flow_layers = []\n", "\n", "    vardeq_layers = [\n", "        CouplingLayer(\n", "            network=GatedConvNet(c_in=2, c_out=2, c_hidden=16),\n", "            mask=create_checkerboard_mask(h=28, w=28, invert=(i % 2 == 1)),\n", "            c_in=1,\n", "        )\n", "        for i in range(4)\n", "    ]\n", "    flow_layers += [VariationalDequantization(vardeq_layers)]\n", "\n", "    flow_layers += [\n", "        CouplingLayer(\n", "            network=GatedConvNet(c_in=1, c_hidden=32),\n", "            mask=create_checkerboard_mask(h=28, w=28, invert=(i % 2 == 1)),\n", "            c_in=1,\n", "        )\n", "        for i in range(2)\n", "    ]\n", "    flow_layers += [SqueezeFlow()]\n", "    for i in range(2):\n", "        flow_layers += [\n", "            CouplingLayer(\n", "                network=GatedConvNet(c_in=4, c_hidden=48), mask=create_channel_mask(c_in=4, invert=(i % 2 == 1)), c_in=4\n", "            )\n", "        ]\n", "    flow_layers += [SplitFlow(), SqueezeFlow()]\n", "    for i in range(4):\n", "        flow_layers += [\n", "            CouplingLayer(\n", "                network=GatedConvNet(c_in=8, c_hidden=64), mask=create_channel_mask(c_in=8, invert=(i % 2 == 1)), c_in=8\n", "            )\n", "        ]\n", "\n", "    flow_model = ImageFlow(flow_layers).to(device)\n", "    return flow_model"]}, {"cell_type": "markdown", "id": "72452657", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.018848, "end_time": "2022-05-12T12:19:12.352716", "exception": false, "start_time": "2022-05-12T12:19:12.333868", "status": "completed"}, "tags": []}, "source": ["We can show the difference in number of parameters below:"]}, {"cell_type": "code", "execution_count": 23, "id": "770b9376", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:12.393479Z", "iopub.status.busy": "2022-05-12T12:19:12.393142Z", "iopub.status.idle": "2022-05-12T12:19:16.320589Z", "shell.execute_reply": "2022-05-12T12:19:16.319851Z"}, "papermill": {"duration": 3.94926, "end_time": "2022-05-12T12:19:16.322228", "exception": false, "start_time": "2022-05-12T12:19:12.372968", "status": "completed"}, "tags": []}, "outputs": [{"name": "stdout", "output_type": "stream", "text": ["Number of parameters: 335,128\n", "Number of parameters: 379,556\n", "Number of parameters: 1,062,090\n"]}], "source": ["def print_num_params(model):\n", "    num_params = sum(np.prod(p.shape) for p in model.parameters())\n", "    print(f\"Number of parameters: {num_params:,}\")\n", "\n", "\n", "print_num_params(create_simple_flow(use_vardeq=False))\n", "print_num_params(create_simple_flow(use_vardeq=True))\n", "print_num_params(create_multiscale_flow())"]}, {"cell_type": "markdown", "id": "cbc80813", "metadata": {"papermill": {"duration": 0.018612, "end_time": "2022-05-12T12:19:16.360971", "exception": false, "start_time": "2022-05-12T12:19:16.342359", "status": "completed"}, "tags": []}, "source": ["Although the multi-scale flow has almost 3 times the parameters of the single scale flow,\n", "it is not necessarily more computationally expensive than its counterpart.\n", "We will compare the runtime in the following experiments as well."]}, {"cell_type": "markdown", "id": "c7456b66", "metadata": {"papermill": {"duration": 0.018628, "end_time": "2022-05-12T12:19:16.399001", "exception": false, "start_time": "2022-05-12T12:19:16.380373", "status": "completed"}, "tags": []}, "source": ["## Analysing the flows\n", "\n", "In the last part of the notebook, we will train all the models we have implemented above,\n", "and try to analyze the effect of the multi-scale architecture and variational dequantization.\n", "\n", "### Training flow variants\n", "\n", "Before we can analyse the flow models, we need to train them first.\n", "We provide pre-trained models that contain the validation and test performance, and run-time information.\n", "As flow models are computationally expensive, we advice you to rely on\n", "those pretrained models for a first run through the notebook."]}, {"cell_type": "code", "execution_count": 24, "id": "f7543171", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:16.439375Z", "iopub.status.busy": "2022-05-12T12:19:16.438832Z", "iopub.status.idle": "2022-05-12T12:19:16.736183Z", "shell.execute_reply": "2022-05-12T12:19:16.735384Z"}, "papermill": {"duration": 0.320435, "end_time": "2022-05-12T12:19:16.738208", "exception": false, "start_time": "2022-05-12T12:19:16.417773", "status": "completed"}, "tags": []}, "outputs": [{"name": "stderr", "output_type": "stream", "text": ["GPU available: True, used: True\n"]}, {"name": "stderr", "output_type": "stream", "text": ["TPU available: False, using: 0 TPU cores\n"]}, {"name": "stderr", "output_type": "stream", "text": ["IPU available: False, using: 0 IPUs\n"]}, {"name": "stderr", "output_type": "stream", "text": ["HPU available: False, using: 0 HPUs\n"]}, {"name": "stderr", "output_type": "stream", "text": ["GPU available: True, used: True\n"]}, {"name": "stderr", "output_type": "stream", "text": ["TPU available: False, using: 0 TPU cores\n"]}, {"name": "stderr", "output_type": "stream", "text": ["IPU available: False, using: 0 IPUs\n"]}, {"name": "stderr", "output_type": "stream", "text": ["HPU available: False, using: 0 HPUs\n"]}, {"name": "stderr", "output_type": "stream", "text": ["GPU available: True, used: True\n"]}, {"name": "stderr", "output_type": "stream", "text": ["TPU available: False, using: 0 TPU cores\n"]}, {"name": "stderr", "output_type": "stream", "text": ["IPU available: False, using: 0 IPUs\n"]}, {"name": "stderr", "output_type": "stream", "text": ["HPU available: False, using: 0 HPUs\n"]}, {"name": "stdout", "output_type": "stream", "text": ["Found pretrained model, loading...\n", "Found pretrained model, loading...\n", "Found pretrained model, loading...\n"]}], "source": ["flow_dict = {\"simple\": {}, \"vardeq\": {}, \"multiscale\": {}}\n", "flow_dict[\"simple\"][\"model\"], flow_dict[\"simple\"][\"result\"] = train_flow(\n", "    create_simple_flow(use_vardeq=False), model_name=\"MNISTFlow_simple\"\n", ")\n", "flow_dict[\"vardeq\"][\"model\"], flow_dict[\"vardeq\"][\"result\"] = train_flow(\n", "    create_simple_flow(use_vardeq=True), model_name=\"MNISTFlow_vardeq\"\n", ")\n", "flow_dict[\"multiscale\"][\"model\"], flow_dict[\"multiscale\"][\"result\"] = train_flow(\n", "    create_multiscale_flow(), model_name=\"MNISTFlow_multiscale\"\n", ")"]}, {"cell_type": "markdown", "id": "5d40ee0c", "metadata": {"papermill": {"duration": 0.018656, "end_time": "2022-05-12T12:19:16.777476", "exception": false, "start_time": "2022-05-12T12:19:16.758820", "status": "completed"}, "tags": []}, "source": ["### Density modeling and sampling\n", "\n", "Firstly, we can compare the models on their quantitative results.\n", "The following table shows all important statistics.\n", "The inference time specifies the time needed to determine the\n", "probability for a batch of 64 images for each model, and the sampling\n", "time the duration it took to sample a batch of 64 images."]}, {"cell_type": "code", "execution_count": 25, "id": "cdee5b41", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:16.817610Z", "iopub.status.busy": "2022-05-12T12:19:16.817150Z", "iopub.status.idle": "2022-05-12T12:19:16.822903Z", "shell.execute_reply": "2022-05-12T12:19:16.822265Z"}, "papermill": {"duration": 0.027622, "end_time": "2022-05-12T12:19:16.824400", "exception": false, "start_time": "2022-05-12T12:19:16.796778", "status": "completed"}, "tags": []}, "outputs": [{"data": {"text/html": ["<!-- Some HTML code to increase font size in the following table -->\n", "<style>\n", "th {font-size: 120%;}\n", "td {font-size: 120%;}\n", "</style>\n"], "text/plain": ["<IPython.core.display.HTML object>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["%%html\n", "<!-- Some HTML code to increase font size in the following table -->\n", "<style>\n", "th {font-size: 120%;}\n", "td {font-size: 120%;}\n", "</style>"]}, {"cell_type": "code", "execution_count": 26, "id": "1558a7b3", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:16.863872Z", "iopub.status.busy": "2022-05-12T12:19:16.863266Z", "iopub.status.idle": "2022-05-12T12:19:16.882407Z", "shell.execute_reply": "2022-05-12T12:19:16.881767Z"}, "papermill": {"duration": 0.040403, "end_time": "2022-05-12T12:19:16.883868", "exception": false, "start_time": "2022-05-12T12:19:16.843465", "status": "completed"}, "tags": []}, "outputs": [{"data": {"text/html": ["<table>\n", "<thead>\n", "<tr><th>Model     </th><th>Validation Bpd  </th><th>Test Bpd  </th><th>Inference time  </th><th>Sampling time  </th><th>Num Parameters  </th></tr>\n", "</thead>\n", "<tbody>\n", "<tr><td>simple    </td><td>1.109 bpd       </td><td>1.107 bpd </td><td>51 ms           </td><td>50 ms          </td><td>335,128         </td></tr>\n", "<tr><td>vardeq    </td><td>1.068 bpd       </td><td>1.066 bpd </td><td>69 ms           </td><td>50 ms          </td><td>379,556         </td></tr>\n", "<tr><td>multiscale</td><td>1.029 bpd       </td><td>1.026 bpd </td><td>40 ms           </td><td>22 ms          </td><td>1,062,090       </td></tr>\n", "</tbody>\n", "</table>"], "text/plain": ["<IPython.core.display.HTML object>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["\n", "table = [\n", "    [\n", "        key,\n", "        \"%4.3f bpd\" % flow_dict[key][\"result\"][\"val\"][0][\"test_bpd\"],\n", "        \"%4.3f bpd\" % flow_dict[key][\"result\"][\"test\"][0][\"test_bpd\"],\n", "        \"%2.0f ms\" % (1000 * flow_dict[key][\"result\"][\"time\"]),\n", "        \"%2.0f ms\" % (1000 * flow_dict[key][\"result\"].get(\"samp_time\", 0)),\n", "        \"{:,}\".format(sum(np.prod(p.shape) for p in flow_dict[key][\"model\"].parameters())),\n", "    ]\n", "    for key in flow_dict\n", "]\n", "display(\n", "    HTML(\n", "        tabulate.tabulate(\n", "            table,\n", "            tablefmt=\"html\",\n", "            headers=[\"Model\", \"Validation Bpd\", \"Test Bpd\", \"Inference time\", \"Sampling time\", \"Num Parameters\"],\n", "        )\n", "    )\n", ")"]}, {"cell_type": "markdown", "id": "ebfd4803", "metadata": {"papermill": {"duration": 0.019133, "end_time": "2022-05-12T12:19:16.922751", "exception": false, "start_time": "2022-05-12T12:19:16.903618", "status": "completed"}, "tags": []}, "source": ["As we have intially expected, using variational dequantization improves upon standard dequantization in terms of bits per dimension.\n", "Although the difference with 0.04bpd doesn't seem impressive first, it is a considerably step for generative models\n", "(most state-of-the-art models improve upon previous models in a range of 0.02-0.1bpd on CIFAR with three times as high bpd).\n", "While it takes longer to evaluate the probability of an image due to the variational dequantization,\n", "which also leads to a longer training time, it does not have an effect on the sampling time.\n", "This is because inverting variational dequantization is the same as dequantization: finding the next lower integer.\n", "\n", "When we compare the two models to multi-scale architecture, we can see that the bits per dimension score again dropped by about 0.04bpd.\n", "Additionally, the inference time and sampling time improved notably despite having more parameters.\n", "Thus, we see that the multi-scale flow is not only stronger for density modeling, but also more efficient.\n", "\n", "Next, we can test the sampling quality of the models.\n", "We should note that the samples for variational dequantization and standard dequantization are very similar,\n", "and hence we visualize here only the ones for variational dequantization and the multi-scale model.\n", "However, feel free to also test out the `\"simple\"` model.\n", "The seeds are set to obtain reproducable generations and are not cherry picked."]}, {"cell_type": "code", "execution_count": 27, "id": "897b3aa5", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:16.962505Z", "iopub.status.busy": "2022-05-12T12:19:16.961837Z", "iopub.status.idle": "2022-05-12T12:19:17.145249Z", "shell.execute_reply": "2022-05-12T12:19:17.144514Z"}, "papermill": {"duration": 0.205115, "end_time": "2022-05-12T12:19:17.146940", "exception": false, "start_time": "2022-05-12T12:19:16.941825", "status": "completed"}, "tags": []}, "outputs": [{"name": "stderr", "output_type": "stream", "text": ["Global seed set to 44\n"]}, {"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"347.04pt\" height=\"347.04pt\" viewBox=\"0 0 347.04 347.04\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:17.071922</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 347.04 \n", "L 347.04 347.04 \n", "L 347.04 -0 \n", "L 0 -0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g clip-path=\"url(#pc370c39444)\">\n", "    <image xlink:href=\"data:image/png;base64,\n", "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\" id=\"image511f9fa543\" transform=\"scale(1 -1)translate(0 -333.36)\" x=\"7.2\" y=\"-6.48\" width=\"332.64\" height=\"333.36\"/>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"pc370c39444\">\n", "   <rect x=\"7.2\" y=\"7.2\" width=\"332.64\" height=\"332.64\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 600x600 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["pl.seed_everything(44)\n", "samples = flow_dict[\"vardeq\"][\"model\"].sample(img_shape=[16, 1, 28, 28])\n", "show_imgs(samples.cpu())"]}, {"cell_type": "code", "execution_count": 28, "id": "99b4f488", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:17.191393Z", "iopub.status.busy": "2022-05-12T12:19:17.190645Z", "iopub.status.idle": "2022-05-12T12:19:17.331569Z", "shell.execute_reply": "2022-05-12T12:19:17.330846Z"}, "papermill": {"duration": 0.164542, "end_time": "2022-05-12T12:19:17.333213", "exception": false, "start_time": "2022-05-12T12:19:17.168671", "status": "completed"}, "tags": []}, "outputs": [{"name": "stderr", "output_type": "stream", "text": ["Global seed set to 44\n"]}, {"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"347.04pt\" height=\"347.04pt\" viewBox=\"0 0 347.04 347.04\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:17.265579</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 347.04 \n", "L 347.04 347.04 \n", "L 347.04 -0 \n", "L 0 -0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g clip-path=\"url(#p358e99ac8b)\">\n", "    <image xlink:href=\"data:image/png;base64,\n", "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\" id=\"imagec893e6e5c9\" transform=\"scale(1 -1)translate(0 -333.36)\" x=\"7.2\" y=\"-6.48\" width=\"332.64\" height=\"333.36\"/>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"p358e99ac8b\">\n", "   <rect x=\"7.2\" y=\"7.2\" width=\"332.64\" height=\"332.64\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 600x600 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["pl.seed_everything(44)\n", "samples = flow_dict[\"multiscale\"][\"model\"].sample(img_shape=[16, 8, 7, 7])\n", "show_imgs(samples.cpu())"]}, {"cell_type": "markdown", "id": "a54b61c2", "metadata": {"papermill": {"duration": 0.020193, "end_time": "2022-05-12T12:19:17.375862", "exception": false, "start_time": "2022-05-12T12:19:17.355669", "status": "completed"}, "tags": []}, "source": ["From the few samples, we can see a clear difference between the simple and the multi-scale model.\n", "The single-scale model has only learned local, small correlations while the multi-scale model was able to learn full,\n", "global relations that form digits.\n", "This show-cases another benefit of the multi-scale model.\n", "In contrast to VAEs, the outputs are sharp as normalizing flows can naturally model complex,\n", "multi-modal distributions while VAEs have the independent decoder output noise.\n", "Nevertheless, the samples from this flow are far from perfect as not all samples show true digits."]}, {"cell_type": "markdown", "id": "66d1bbfd", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.020133, "end_time": "2022-05-12T12:19:17.417021", "exception": false, "start_time": "2022-05-12T12:19:17.396888", "status": "completed"}, "tags": []}, "source": ["### Interpolation in latent space\n", "\n", "Another popular test for the smoothness of the latent space of generative models is to interpolate between two training examples.\n", "As normalizing flows are strictly invertible, we can guarantee that any image is represented in the latent space.\n", "We again compare the variational dequantization model with the multi-scale model below."]}, {"cell_type": "code", "execution_count": 29, "id": "a7cac8db", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:17.458920Z", "iopub.status.busy": "2022-05-12T12:19:17.458467Z", "iopub.status.idle": "2022-05-12T12:19:17.500905Z", "shell.execute_reply": "2022-05-12T12:19:17.500200Z"}, "papermill": {"duration": 0.065572, "end_time": "2022-05-12T12:19:17.502663", "exception": false, "start_time": "2022-05-12T12:19:17.437091", "status": "completed"}, "tags": []}, "outputs": [], "source": ["@torch.no_grad()\n", "def interpolate(model, img1, img2, num_steps=8):\n", "    \"\"\"\n", "    Args:\n", "        model: object of ImageFlow class that represents the (trained) flow model\n", "        img1, img2: Image tensors of shape [1, 28, 28]. Images between which should be interpolated.\n", "        num_steps: Number of interpolation steps. 8 interpolation steps mean 6 intermediate pictures besides img1 and img2\n", "    \"\"\"\n", "    imgs = torch.stack([img1, img2], dim=0).to(model.device)\n", "    z, _ = model.encode(imgs)\n", "    alpha = torch.linspace(0, 1, steps=num_steps, device=z.device).view(-1, 1, 1, 1)\n", "    interpolations = z[0:1] * alpha + z[1:2] * (1 - alpha)\n", "    interp_imgs = model.sample(interpolations.shape[:1] + imgs.shape[1:], z_init=interpolations)\n", "    show_imgs(interp_imgs, row_size=8)\n", "\n", "\n", "exmp_imgs, _ = next(iter(train_loader))"]}, {"cell_type": "code", "execution_count": 30, "id": "52757b12", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:17.545009Z", "iopub.status.busy": "2022-05-12T12:19:17.544482Z", "iopub.status.idle": "2022-05-12T12:19:17.775782Z", "shell.execute_reply": "2022-05-12T12:19:17.775084Z"}, "papermill": {"duration": 0.253822, "end_time": "2022-05-12T12:19:17.777408", "exception": false, "start_time": "2022-05-12T12:19:17.523586", "status": "completed"}, "tags": []}, "outputs": [{"name": "stderr", "output_type": "stream", "text": ["Global seed set to 42\n"]}, {"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"643.2975pt\" height=\"97.56pt\" viewBox=\"0 0 643.2975 97.56\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:17.639096</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 97.56 \n", "L 643.2975 97.56 \n", "L 643.2975 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g clip-path=\"url(#p23ea946160)\">\n", "    <image xlink:href=\"data:image/png;base64,\n", "iVBORw0KGgoAAAANSUhEUgAAA2oAAAB0CAYAAAD0H8F9AAAdWklEQVR4nO3de7zNVf748VUSIoNBMrlm3O80hsJI7mQkuSQG0ZQMZTDGnRIig0gZVJJ75EgpDeU6uaQHQxGVezeDDLnV94/f77Fa7zfnc/beZ1/WPl7Pv97vx/ucz2fZn73P3h97vde6bvjw4T8bAAAAAIA3rk/0AAAAAAAAEjdqAAAAAOAZbtQAAAAAwDPcqAEAAACAZ7hRAwAAAADPcKMGAAAAAJ65Iag4fPjwOA0DoUrrmnDN/BN0Tbhe/uF6JRf+JiYfXmPJheuVXPibmHyCrgnfqAEAAACAZ7hRAwAAAADPcKMGAAAAAJ7hRg0AAAAAPMONGgAAAAB4hhs1AAAAAPAMN2oAAAAA4JnAfdR8U758eZF/8sknqf7swYMHRV6/fn0bHzhwILoDQ6pq1apl4/fff1/UMmfOLPITJ07YuFmzZqK2ZcuWGIwOWrly5USekpJi40KFConavHnzbDx16lRR+/e//x2D0UHLlSuXyLt27SryggUL2vj48eOiNn78+JiNCwAApB/fqAEAAACAZ7hRAwAAAADPJNXUx59++knkly9fFnmmTJlsfNttt4la3759bTx69GhRO3LkSLSGCOU3v/mNjfVURy1Pnjw21tOyOnfubOMvv/wyOoODyZcvn8iXLl0q8sKFC6f6ux06dLBxu3btRG3+/Pki79Wrl41PnToV9jjxi6pVq9p4xowZolapUiWRX3fddTa+dOmSqLmvqeeff17UXnzxxXSPE/+PO/3UGGOaNGli4xdeeEHU3PcwY4x5++23bayntX7zzTfRGiIAwFN8owYAAAAAnuFGDQAAAAA8w40aAAAAAHgmqXrUdu/eLfJFixaJ3O2TWb9+vaj17NkzdgNDqtwl+VevXi1q99xzj8jdHsS//OUvokZfWniyZctm41KlSola7dq1bay3QShevHhE53N7oa523A0bNtiY/qcrXX+9/D8z9zq4/X3GGNOpUycb33zzzYHH/fnnn22s+59Kly5t43Hjxola9erVRe6O4ccffww857XI7cXt3bu3qLVp00bkbt+ne32upnHjxjZesWKFqG3dutXGr732mqht3LgxjRFf2264QX70qVevno2LFSsmatOmTQv5uLq3+ocffrDx9u3bRc3dXujo0aMhnwNXvt80aNDAxlmyZBE1/ffKfW88e/asqLmfV9J6bQLxwjdqAAAAAOAZbtQAAAAAwDNJNfVRmzBhgsjdqY8lSpQQtcqVK9t4x44dsRwWHCdOnLCxnp7jTlcwRk7/atGihai500RgTIECBURerVo1kbtTR+vXry9q7rQRPb1Db4HhXhM93SRoasjatWtFPnbsWBvXqFFD1Lp37y5yve3GtaBPnz4id6cihvO4a0HX2q1lz55d1Lp06SJy9+/pkiVLRE0v7X8t0NfrySeftLE7DdKYKx/3SK+fuy2Dzrt16yZqeprkG2+8YeO5c+eGfP6MJG/evDaeMmWKqLVt29bG6Zny1q9fv5B/9vTp0zaeOXOmqB08eFDkkydPjnhMGcWoUaNsPHjwYFHT71uRcq+9/rt76NAhkbvXWm85o9s8ojW+ZFKrVi2Rr1u3Lq7n19dLv6e5dKvUxYsXYzKmSPGNGgAAAAB4hhs1AAAAAPAMN2oAAAAA4Jmk7lHT87h37dpl4/Lly4ua2yPTqFGj2A4MV3Xu3LnAujsn/NFHHxW1p556KiZjSlZ6CWm3B8UY+VguXrxY1O6++24b58mTR9T0Eu6uoB4nPQd/5MiRIp89e7aNJ02aJGq63yAlJcXGeknrZOYuC71p0yZR03+vXEE9M7qPIpzfDfo5fVx3Swc91pIlS9pYL/N/+PDhkM7vI72E+7Bhw2zcv39/UQt63aR1jUIV1Kuoz9+yZUuRu1uhdOjQQdSGDBli44z0enP70o0x5uWXX7ZxhQoVRC0eS7Hrc7hbazzxxBOidunSJZHfe++9Nn7mmWdEzV1SPpnlypVL5O+9957IK1WqZONY9XwFvVYLFSok8vnz56f6s3oLGrdndOXKlRGOzm96zQHddxlv+nrpvkGXXu/iX//6l43feeed6A4sAnyjBgAAAACe4UYNAAAAADzDjRoAAAAAeCape9S+++47kbv7Jug+inLlytn49ttvF7X9+/fHYHTQjhw5InI9D9/tCfn1r38tam5f4apVq0I+5/Tp00X+zTff2FjPWf7www9DPm6iff311yLXPSpur0nnzp1FLV++fDbWe5aNHj1a5A899JCNg+bvu/utGWNMjx49RN63b18buz0xxhjz6quvinzQoEE2Xrp0qai5vRxHjx5NdTw+ch+joJ60cKTVWxN0zYL2DArqh8qdO7eo9ezZ08a6/3fhwoUid/dc089h3+i95QYMGGBj/XwPklb/XyyOo3/W/bc0btxY1KpUqWLjBQsWiNq0adNE/vnnn6d6Dt+4PWnGXNmXFm/hXC/dH1mvXj0b69479++n7o3Sn230e65PZsyYIXK9b2AyeeSRR0R+33332XjixImi5n5GcffWM8b/11j16tVtrHvS9H6SPnM/nxhjTPv27W3csWNHUdP95RcuXIjdwP4/vlEDAAAAAM9wowYAAAAAnknqqY+aO02qSZMmonbrrbfauEyZMqLG1Mf42Lx5s8h37twpcncKTubMmUXNXR48ramP7jLo7jGNMaZatWo21lNgk2nqY8OGDQPr7rLo58+fT7WmudPYjJFLJrtLRBsTPC3j97//vcjdMejpcO3atRN58+bNbXz//feL2pYtW2w8fvz4VM/vowceeMDGaU1jS8+S/KEeJ4he/vrkyZM2zpkzp6i507RKlCghagMHDhS5OwWoW7duoqb/PiSau42FMaFPdwzn2qbnukdrWtQtt9xi4169eomazlu3bm3jN998Myrnjxb9vl6kSJEEjeTqgp4X4bym9dTj3r17XzU2xphWrVqJfPny5SGfJ95q1qyZ6CHEjNtuoNsL3O0W3G00jJHLxPvoV7/6lY2TaapjWgoWLGhjfQ327Nkjcvc9be/evTEZD9+oAQAAAIBnuFEDAAAAAM9wowYAAAAAnsmwPWqPP/64qFWsWNHG/fr1EzU9B/Xs2bMxGB00vfSzXp7X5S4xr5eB/eKLL0R+7tw5G+u+M7dHTT8Pkom73YQxV/Y4uHPiw6Gf+/nz57dxUI+FrlWqVEnkd9xxh40/+ugjUdNL47pL/rr9M8aEtyy6b8aMGWNj/e+I1hLu0epj0sdxl5v+29/+Jmruayqt85UsWdLG+nWsn9OJ5m4lYEzw8z2oFqvl+RNh7ty5Ns6RI0cCR3Ild1sPY67spUyEUPsRI31OpEUvBZ+SknLV8yeKuwR/pO9Z0eR+lnD7oY25sv/WfY/T2+OEw70O+nrp91Hf6G0I4sHtl3766adT/bm8efOK3N1eJT10L6y7fP/QoUOjcg4teT/1AAAAAEAGxY0aAAAAAHiGGzUAAAAA8EyG6lE7ceKEjXV/wUsvvWTju+66S9T0fjkrVqyIweigrV69WuTHjh2zsbvvnTFyX4umTZuK2tSpU1M9x+zZs0X+5JNP2jg988oTLa29/9w9TdzHzhhjjh49GvJ55syZY2M9Xz5r1qwhHyeoR23fvn0i37Fjh40bNWokahUqVAj5nL7ZtGmTjd294tKSnh6noN6pcJw5c8bGeh+gxYsXRzSem266KeLxxIPufS1QoICN09MLGNSbFOm+aum5tuHs3ebuUfmHP/xB1NauXRvxGKLB3dMpGYRzzSLtYStatKjIW7ZsaeNly5ZFdMxocp9P7n6MieK+xm+77TZRc/fLMsaYZ5991sZ6v89Ir5fbw2uM3HvTmCv3IE00998dr55H933jwQcfFLVx48bZeN68eaKmP28OGTLExnXq1Il4PA8//LCN3R5QY67sc4wU36gBAAAAgGe4UQMAAAAAzyT+u+YY2bZtm8hPnz5tY71s75/+9CeRM/UxPg4ePChyd5qdXkrV/Vq9e/fuohY09fH8+fMid6ck6KkNhw4dSmPE/tBf67tTWoyRX+W3aNFC1F588cWQzzN9+nQb58qVS9RGjhxp47SmkbrTDNzrbIx8bRpjzKlTp1I9jp7Kk0ymTJli4zvvvFPU8uTJI/ILFy7YeMSIEaLmXs/cuXOLmp464z7f07ME+IQJE2zsTlHWx01r+os7nVBPF/LN5MmTRe5u8ZI9e/aQjxM0nTGtxyvUa5aeqZhB5wg6zuXLl0M+Rzy88847ItfTphMx1T3osXXHE69pY3rJ8kTT78+J5k7FrFWrlqh99dVXIo/FNbvxxhtFrt9zIR+jypUri9rrr78e59HILYRitWUJ36gBAAAAgGe4UQMAAAAAz3CjBgAAAACeybA9au4S38YY07VrVxvPmjVL1PLnzy9yd56puyw1YmvgwIE2rlu3rqj97ne/s3GRIkVErVq1aiJ3+xMvXbokam7/0+effx75YBPs66+/Frn72BljzLvvvmtjdw51eowZM0bkq1atsvHy5ctFTW+v4PZGjB8/XtQee+wxkbuvXb08sW89FuF4//33baz7KgcPHixyt7flgw8+EDX3OuheKfd1YozcUkH3WCxatCiUYRtjjClbtqyNy5QpI2rh9Gq4Y/jkk09C/r1E0Eth9+7d28Y1atQQtWgtla8FLeWf2s+l9bNp/W6ognpJE0G/poYPHy5y3QcaC+Fch59++imknwvnHGmdPz19qrGwdetWG+v36ngs1x/O9YpXH2Gizwn/8I0aAAAAAHiGGzUAAAAA8EyGnfqoLV261Mb9+/cXNb1U9rBhw2zcr1+/2A4MV9WuXTuRHzhwwMZ6e4VJkyaJvF69ejbW07327t1r46xZs6Z7nL7YvHmzyN3l+/v06SNq7hSh77//PuJzfvzxxzZ+7733RK1z586p/p47DdkYY3bt2iXyzz77LNXfLVGihI3vuusuUVu/fn3qg/XMggULRK63CClUqJCN3a0NjJFbMfzvf/8TtTVr1qSaly5dOuTxRTpFKq1l4gsXLhzRcX2wfft2G+upj0FiNX0p1GmR0RyDu8XEp59+GpVjxsq+fftEHs41c4UzPS6c100sptnpY+q/7z5P99+zZ4/IK1SoENFx0rpe4bxu4u3kyZMiP3LkSGIGEiL3+ZWeqcXuNGC9/YveoiCcrVHi4YcffrBxrLab4Bs1AAAAAPAMN2oAAAAA4Blu1AAAAADAM9dMj1o4unTpYuPJkyeL2qFDh+I9nGvS0aNHRb569WobN2jQQNRq1qwp8unTp9tYL3u+cuVKG+v+nowkJSXFxt26dRO1Dh062NjtOUkP3R8W1KOm+wImTpwY0Tn1FgXJRPdj/POf/xT5iBEjbNywYUNRmz17to3dviljjHnppZdE7m4vcvHiRVGLVh9aOD/rW09IOMaOHWtjd9sDY67sc3ZFa+n19Bw31J9NqzfKfe5duHAh5PMngu5F19tchCoRz9loba/wzDPPiFz3sPrEXRvAGGPeeOONiI6TVp+sb9z+rGnTpoma+3nFR+5aAu6WQOE6e/asjTt27ChqxYsXF3nTpk1t3Lp164jPGalvv/1W5D179rTxxo0bY3JOvlEDAAAAAM9wowYAAAAAnuFGDQAAAAA8c032qA0fPlzkr7/+ushz585t46pVq4oaPWrxcfnyZZGvWrXKxnfffbeoZcqUSeS1a9e2cZYsWUTN7f3JyNx9zQ4fPixq48ePt7HuJduyZYvIZ82aZeNt27aJmju3Pm/evKLm7i1ijDE5cuSwcXp6CNw9xfbv3x/y7/luyZIlInefp/rxat++/VVjY4ypWLGiyN2elHz58oU8nqBeqXBqWubMmUMeg2/c11G09jeK1p5P6TlO0DXTP6t7h3126tQpkZ87d87G2bJli/dwrhDO371QX2N6H6eFCxeGP7AE+fHHH+N+zqDXjf4Monui161bZ+NSpUqJWuXKlQPP45o/f76Nhw4dGjzgDMr9fOD2wRpzZY+9u4bEW2+9FZXz6159t+de73mqe9Ti0UfIN2oAAAAA4Blu1AAAAADAM95PfXSXPa5Ro4ao9ejRI9Xfmzt3rsg3bdpkY3canTFyKXNjjHnooYds/Morr4ha2bJlbZxM00BipXHjxiK/9957RV6uXDkb6yWt9fS4HTt22FgvI+wuW3vPPfcEjuH222+38SOPPCJqAwcONNcCdwrM5s2bRa1NmzY2rlKliqjpKRvua0xP33CnhugpbfraBglnKtbx48dt7E69THb6MXCX0g9nuqD7t+tqeWritSx8rJYvjjd3Gp0WzWXzQ53umJ7jhDOlUk8z99nOnTtF7m7V8txzz4laPJZwD3qN6b9lkT6H3C0kjIneFN140O/5etuWP//5zzYOZ+pqONOA3WXi9WeZY8eOiXzfvn02vuWWW0TtvvvuE/kXX3xh42LFiomau6WK71sJaBs2bLDxqFGjRM3dBsgY+bksSNGiRUW+ePFikbuPZfPmzUUt0i173O2fjJFbN508eTKiY0YT36gBAAAAgGe4UQMAAAAAz3CjBgAAAACeSXiP2k033STy1157TeRuP1L27NlDPq5egt9d+tWdV2uMMZ999lmqx7n55ptF3qdPHxv3798/5PFkJI0aNbLxihUrRC2cufX6sS1YsKCNmzRpImoNGza0cZkyZUI+R69evUTubsWgexgyqnD6FMLpZbnhhl/+fOi59e5yu+EKGoNeMjmj0PPg3X6IwoULx3k0kUvr9e8udXz99fL/CZOp5/Dpp58WeadOnWwczt/A9IjFefTrWOfhvAf75vnnn7exXrp/5syZNvaxTyhoTG6tVatWorZ7926R634fn1y4cEHkf/3rX0XuLoPubj8TrqDXjft5dMaMGaI2ZswYkbu9Unq9Ave5lpG5n6v1Z+5x48aJPJy+dZe7XZbO9bY2c+bMsbG+l3D7zjTfezn5Rg0AAAAAPMONGgAAAAB4hhs1AAAAAPBMwnvU9D4ILVu2TPVnT5w4IfKFCxeKvGnTpjbWfR1Zs2a1cf369UVN50Fq1qxpY91f5+7BkZHoXpLHH3/cxuH0SejeQHePPE0ft0GDBiGfx+Ved2MinyedbNz+P7evMlxu35B+HsRDWj0zGYXuLYm0Ly2c/dAi+bn0nn/RokU2/vvf/y5q48ePj8oY4qFIkSIij/Txi/R6RVM4+7G5PU/6b6vv3Pf5oUOHRnyccB4vV9DPhvO3NehvYIUKFUQ+aNAgkfvco5aWUqVK2TicvTcjfY3p/c6mT58u8u7du9t4/fr1ota3b99Uj1u3bl2Ru3vE6uszbNiwkMZ6rXI/j+tcr3Pg9tMZY8w//vEPG+u9Zn3DN2oAAAAA4Blu1AAAAADAMwmf+rhu3bqQf1Yv1/rKK6+IPGfOnDbOnDmzqBUvXtzGHTp0ELX27duLPF++fKmOoVatWjbWy5Fm1OX69bLZ7uPuTje9GneZWn399BSAS5cu2bh06dKi1rZtWxvXqVMnjRH/Qk9zmD17to3r1asX8nGSzZkzZ2w8depUUevZs2eqv6enicydO9fGLVq0ELVcuXKlY4Sh0dfPfY1nJHoad9euXW1cuXLlkI+TiKlz4ZzffX4dP3481sOJmS1btoh8zZo1Ng7n70qir1e4Ywha4tp3q1atsrFePn3kyJE21lM6k2l6qh6rXjY+mX3wwQc21lM6hwwZYuMbb7xR1KJ1vfRxqlevbuOqVauK2ttvvy1ydxpeyZIlRS1v3rw2Xr58ebrH6Ytz586JPH/+/DZ+7LHHRG3AgAE2zpYtW1TOrz+vaI0bN7Zxs2bNRC2oRWbXrl0ij8e2MnyjBgAAAACe4UYNAAAAADzDjRoAAAAAeCbhPWp6zvvOnTtF7i43q+e8aqdPn0619v3339tY9xfMmTNH5A888ICN3SVYjTHm4MGDNt64cWPgeDIq9zqkNX/fne9/8eJFUVu9enWq51i7dq3I33zzTRv/8Y9/FLXRo0eLPKiP6dZbb021lpG412X//v0h/56+fu6cebcPwBhj7r//fht//PHHotamTRuRu3P09dLm4cioy/N/++23Ip8yZYqNZ86cGe/hxExKSoqN3f7HZKN7GNytBnSfSVDPs+8uX74s8okTJyZoJNE1YcIEkbt9TaNGjRK1oKXg49WvFul5li1bFt2BJJC7NYQbGyM/k+jPA7HqWXPp7RXeffddkSfiOZNo+r3a/QyuX2Pu0vl6DYK01kGIlLuF0Ycffhjy73Xq1Enk8Xgf4xs1AAAAAPAMN2oAAAAA4JmET33U0xXd3cKNkcup165dW9T0ktaR2r59e6r50KFDRc1ditNdTv5a8tZbb9l4/vz5ovbggw+KXE93jNSxY8ds/MILL4jap59+KvIRI0bY+M477xS1cKYBZhT68SpatKiN+/TpI2p6qVl3+sKhQ4dEbdKkSan+np666m4JMGbMGFELZzneaD2ffOdOx3af+8YY89xzz9lYb2MRNDVUT89xfzZWS5DrqXPjxo1L9ZzJ7KOPPrJx8+bNRU0vBX/HHXeEfFz3OiTi8frPf/4j8qeeeiruY4gH92/SgQMHRM2d1mqMMeXLl0/1OEHTJOPhq6++EvnKlSvjev5EcT836mlskydPFnnNmjVjPp6g665rmzZtsrFu/blWPPvsszbWUwlnzJghcndZ/UR49NFHRc7URwAAAAC4BnGjBgAAAACe4UYNAAAAADyT8B417eWXXw7M4+3ChQsJPb/vOnbsGJjHw5o1a0S+Z88eG7dt21bU9HYQ1wL9HH7iiSeuGodL96UFmTp1qo11j0C7du1S/T3d4/Tf//435HMmM/ffvWrVKlGrWLGijYsXLy5qOnf7odzXhTFySWTd/6v7rMqUKRPKsK+wfv16ke/YsSOi4ySTrVu3irxu3boir169uo2bNWsmag8//LDI3cdL97YULFjQxr/97W9FLXPmzKEPOMCuXbuichzfuY/tggULRG3z5s0i79Gjh427dOkiam6vojHG3HDDLx+x9LYk7msqWj2hp06dEnmWLFmictxkotcc6Ny5s8gHDBhg471794pa4cKFbVygQAFRa926dbSGKJQtW9bG7nZUxhizbdu2mJzTZ0ePHhV5q1atRO6+pho2bChq/fv3t3GVKlVETW/TECn3ehljTNeuXW08a9asqJxD4xs1AAAAAPAMN2oAAAAA4Blu1AAAAADAM971qAHpdfz4cRu7e33BD4MHDxa53t/L3ectJSVF1JYsWRKzcSULt39t3759oqZz3d+WGndvRGNkT6Exch6+2+NhjOyD+e6770RN7zlzLfb8nj9/XuQbNmy4amzMlXt2hUr3eXbo0EHk7t5DmTJlSvU4es++6dOnRzSejETvTTZo0KCrxuFy96Rr0qSJqFWuXDnk47j9daNHjxY1vSfctUjvner2GAbRfYP9+vUTefv27W3s9g2Hy92H61rsSUuLfs9w82XLlomam+v3qTp16og80v3Ydu/eLfJY9aW5+EYNAAAAADzDjRoAAAAAeIapjwDi6ssvvxS5nqaFxDt06JDIR4wYcdUYfpg/f35g7m7T0KJFC1GrVauWjRcuXChqmzZtitYQobhTwMeMGSNqxYoVE/ns2bNtXLVqVVFzpz7qabaInN4OY9y4cSJ/9dVXbVypUiVRmzdvnshz5cplYz2lUk/9R3SMHTtW5O4UU2OMKVWqlI2XLl0qajly5LCxvl7R2kojHDxDAAAAAMAz3KgBAAAAgGe4UQMAAAAAz9CjBgBABrZly5arxvDDmTNnRL5z506RV69ePZ7DQQjcbYDc2Bhj8uTJE+/hIA2HDx9ONc+ZM2e8hxMWvlEDAAAAAM9wowYAAAAAnuFGDQAAAAA8w40aAAAAAHiGGzUAAAAA8Aw3agAAAADgGW7UAAAAAMAz3KgBAAAAgGe4UQMAAAAAz3CjBgAAAACeuW748OE/J3oQAAAAAIBf8I0aAAAAAHiGGzUAAAAA8Aw3agAAAADgGW7UAAAAAMAz3KgBAAAAgGe4UQMAAAAAz/wfTcunkHtJqUoAAAAASUVORK5CYII=\" id=\"image2de3d4d11a\" transform=\"scale(1 -1)translate(0 -83.52)\" x=\"7.2\" y=\"-6.84\" width=\"629.28\" height=\"83.52\"/>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"p23ea946160\">\n", "   <rect x=\"7.2\" y=\"7.2\" width=\"628.8975\" height=\"83.16\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 1200x150 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}, {"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"643.2975pt\" height=\"97.56pt\" viewBox=\"0 0 643.2975 97.56\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:17.736600</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 97.56 \n", "L 643.2975 97.56 \n", "L 643.2975 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g clip-path=\"url(#p5d75e28bc4)\">\n", "    <image xlink:href=\"data:image/png;base64,\n", "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\" id=\"imaged75002c20c\" transform=\"scale(1 -1)translate(0 -83.52)\" x=\"7.2\" y=\"-6.84\" width=\"629.28\" height=\"83.52\"/>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"p5d75e28bc4\">\n", "   <rect x=\"7.2\" y=\"7.2\" width=\"628.8975\" height=\"83.16\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 1200x150 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["pl.seed_everything(42)\n", "for i in range(2):\n", "    interpolate(flow_dict[\"vardeq\"][\"model\"], exmp_imgs[2 * i], exmp_imgs[2 * i + 1])"]}, {"cell_type": "code", "execution_count": 31, "id": "e91953f8", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:17.825477Z", "iopub.status.busy": "2022-05-12T12:19:17.824997Z", "iopub.status.idle": "2022-05-12T12:19:18.074474Z", "shell.execute_reply": "2022-05-12T12:19:18.073638Z"}, "papermill": {"duration": 0.275174, "end_time": "2022-05-12T12:19:18.076116", "exception": false, "start_time": "2022-05-12T12:19:17.800942", "status": "completed"}, "tags": []}, "outputs": [{"name": "stderr", "output_type": "stream", "text": ["Global seed set to 42\n"]}, {"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"643.2975pt\" height=\"97.56pt\" viewBox=\"0 0 643.2975 97.56\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:17.931335</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 97.56 \n", "L 643.2975 97.56 \n", "L 643.2975 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g clip-path=\"url(#pa1616ce8c4)\">\n", "    <image xlink:href=\"data:image/png;base64,\n", "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\" id=\"image8d0d400f6a\" transform=\"scale(1 -1)translate(0 -83.52)\" x=\"7.2\" y=\"-6.84\" width=\"629.28\" height=\"83.52\"/>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"pa1616ce8c4\">\n", "   <rect x=\"7.2\" y=\"7.2\" width=\"628.8975\" height=\"83.16\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 1200x150 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}, {"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"643.2975pt\" height=\"97.56pt\" viewBox=\"0 0 643.2975 97.56\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:18.034217</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 97.56 \n", "L 643.2975 97.56 \n", "L 643.2975 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g clip-path=\"url(#p5f0b2b5e27)\">\n", "    <image xlink:href=\"data:image/png;base64,\n", "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\" id=\"image3739852647\" transform=\"scale(1 -1)translate(0 -83.52)\" x=\"7.2\" y=\"-6.84\" width=\"629.28\" height=\"83.52\"/>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"p5f0b2b5e27\">\n", "   <rect x=\"7.2\" y=\"7.2\" width=\"628.8975\" height=\"83.16\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 1200x150 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["pl.seed_everything(42)\n", "for i in range(2):\n", "    interpolate(flow_dict[\"multiscale\"][\"model\"], exmp_imgs[2 * i], exmp_imgs[2 * i + 1])"]}, {"cell_type": "markdown", "id": "3f492396", "metadata": {"papermill": {"duration": 0.02286, "end_time": "2022-05-12T12:19:18.124072", "exception": false, "start_time": "2022-05-12T12:19:18.101212", "status": "completed"}, "tags": []}, "source": ["The interpolations of the multi-scale model result in more realistic digits\n", "(first row $7\\leftrightarrow 8\\leftrightarrow 6$, second row $9\\leftrightarrow 4\\leftrightarrow 6$),\n", "while the variational dequantization model focuses on local patterns that globally do not form a digit.\n", "For the multi-scale model, we actually did not do the \"true\" interpolation between the two images\n", "as we did not consider the variables that were split along the flow (they have been sampled randomly for all samples).\n", "However, as we will see in the next experiment, the early variables do not effect the overall image much."]}, {"cell_type": "markdown", "id": "702b8862", "metadata": {"papermill": {"duration": 0.022287, "end_time": "2022-05-12T12:19:18.168450", "exception": false, "start_time": "2022-05-12T12:19:18.146163", "status": "completed"}, "tags": []}, "source": ["### Visualization of latents in different levels of multi-scale\n", "\n", "In the following we will focus more on the multi-scale flow.\n", "We want to analyse what information is being stored in the variables split at early layers,\n", "and what information for the final variables.\n", "For this, we sample 8 images where each of them share the same final latent variables,\n", "but differ in the other part of the latent variables.\n", "Below we visualize three examples of this:"]}, {"cell_type": "code", "execution_count": 32, "id": "f5575fa2", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:18.215227Z", "iopub.status.busy": "2022-05-12T12:19:18.214742Z", "iopub.status.idle": "2022-05-12T12:19:18.457363Z", "shell.execute_reply": "2022-05-12T12:19:18.456683Z"}, "papermill": {"duration": 0.267682, "end_time": "2022-05-12T12:19:18.459179", "exception": false, "start_time": "2022-05-12T12:19:18.191497", "status": "completed"}, "tags": []}, "outputs": [{"name": "stderr", "output_type": "stream", "text": ["Global seed set to 44\n"]}, {"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"341.674839pt\" height=\"180.72pt\" viewBox=\"0 0 341.674839 180.72\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:18.261454</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 180.72 \n", "L 341.674839 180.72 \n", "L 341.674839 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g clip-path=\"url(#pcb9b96b23a)\">\n", "    <image xlink:href=\"data:image/png;base64,\n", "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\" id=\"imageddffba6afd\" transform=\"scale(1 -1)translate(0 -166.32)\" x=\"7.2\" y=\"-7.2\" width=\"327.6\" height=\"166.32\"/>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"pcb9b96b23a\">\n", "   <rect x=\"7.2\" y=\"7.2\" width=\"327.274839\" height=\"166.32\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 600x300 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}, {"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"341.674839pt\" height=\"180.72pt\" viewBox=\"0 0 341.674839 180.72\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:18.340864</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 180.72 \n", "L 341.674839 180.72 \n", "L 341.674839 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g clip-path=\"url(#p765ee45c9d)\">\n", "    <image xlink:href=\"data:image/png;base64,\n", "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\" id=\"imagee1300de7b9\" transform=\"scale(1 -1)translate(0 -166.32)\" x=\"7.2\" y=\"-7.2\" width=\"327.6\" height=\"166.32\"/>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"p765ee45c9d\">\n", "   <rect x=\"7.2\" y=\"7.2\" width=\"327.274839\" height=\"166.32\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 600x300 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}, {"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"341.674839pt\" height=\"180.72pt\" viewBox=\"0 0 341.674839 180.72\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:18.417020</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 180.72 \n", "L 341.674839 180.72 \n", "L 341.674839 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g clip-path=\"url(#p9dde5aa4fc)\">\n", "    <image xlink:href=\"data:image/png;base64,\n", "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\" id=\"imageae651c9e35\" transform=\"scale(1 -1)translate(0 -166.32)\" x=\"7.2\" y=\"-7.2\" width=\"327.6\" height=\"166.32\"/>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"p9dde5aa4fc\">\n", "   <rect x=\"7.2\" y=\"7.2\" width=\"327.274839\" height=\"166.32\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 600x300 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["pl.seed_everything(44)\n", "for _ in range(3):\n", "    z_init = flow_dict[\"multiscale\"][\"model\"].prior.sample(sample_shape=[1, 8, 7, 7])\n", "    z_init = z_init.expand(8, -1, -1, -1)\n", "    samples = flow_dict[\"multiscale\"][\"model\"].sample(img_shape=z_init.shape, z_init=z_init)\n", "    show_imgs(samples.cpu())"]}, {"cell_type": "markdown", "id": "01e26db8", "metadata": {"papermill": {"duration": 0.026565, "end_time": "2022-05-12T12:19:18.513844", "exception": false, "start_time": "2022-05-12T12:19:18.487279", "status": "completed"}, "tags": []}, "source": ["We see that the early split variables indeed have a smaller effect on the image.\n", "Still, small differences can be spot when we look carefully at the borders of the digits.\n", "For instance, the hole at the top of the 8 changes for different samples although all of them represent the same coarse structure.\n", "This shows that the flow indeed learns to separate the higher-level\n", "information in the final variables, while the early split ones contain\n", "local noise patterns."]}, {"cell_type": "markdown", "id": "4fc30ffb", "metadata": {"lines_to_next_cell": 2, "papermill": {"duration": 0.022768, "end_time": "2022-05-12T12:19:18.562088", "exception": false, "start_time": "2022-05-12T12:19:18.539320", "status": "completed"}, "tags": []}, "source": ["### Visualizing Dequantization\n", "\n", "As a final part of this notebook, we will look at the effect of variational dequantization.\n", "We have motivated variational dequantization by the issue of sharp edges/boarders being difficult to model,\n", "and a flow would rather prefer smooth, prior-like distributions.\n", "To check how what noise distribution $q(u|x)$ the flows in the\n", "variational dequantization module have learned, we can plot a histogram\n", "of output values from the dequantization and variational dequantization\n", "module."]}, {"cell_type": "code", "execution_count": 33, "id": "d81503b3", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:18.615278Z", "iopub.status.busy": "2022-05-12T12:19:18.613123Z", "iopub.status.idle": "2022-05-12T12:19:18.657172Z", "shell.execute_reply": "2022-05-12T12:19:18.656456Z"}, "papermill": {"duration": 0.071507, "end_time": "2022-05-12T12:19:18.658957", "exception": false, "start_time": "2022-05-12T12:19:18.587450", "status": "completed"}, "tags": []}, "outputs": [], "source": ["def visualize_dequant_distribution(model: ImageFlow, imgs: Tensor, title: str = None):\n", "    \"\"\"\n", "    Args:\n", "        model: The flow of which we want to visualize the dequantization distribution\n", "        imgs: Example training images of which we want to visualize the dequantization distribution\n", "    \"\"\"\n", "    imgs = imgs.to(device)\n", "    ldj = torch.zeros(imgs.shape[0], dtype=torch.float32).to(device)\n", "    with torch.no_grad():\n", "        dequant_vals = []\n", "        for _ in tqdm(range(8), leave=False):\n", "            d, _ = model.flows[0](imgs, ldj, reverse=False)\n", "            dequant_vals.append(d)\n", "        dequant_vals = torch.cat(dequant_vals, dim=0)\n", "    dequant_vals = dequant_vals.view(-1).cpu().numpy()\n", "    sns.set()\n", "    plt.figure(figsize=(10, 3))\n", "    plt.hist(dequant_vals, bins=256, color=to_rgb(\"C0\") + (0.5,), edgecolor=\"C0\", density=True)\n", "    if title is not None:\n", "        plt.title(title)\n", "    plt.show()\n", "    plt.close()\n", "\n", "\n", "sample_imgs, _ = next(iter(train_loader))"]}, {"cell_type": "code", "execution_count": 34, "id": "2df79715", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:18.711236Z", "iopub.status.busy": "2022-05-12T12:19:18.710706Z", "iopub.status.idle": "2022-05-12T12:19:19.802681Z", "shell.execute_reply": "2022-05-12T12:19:19.801929Z"}, "papermill": {"duration": 1.121703, "end_time": "2022-05-12T12:19:19.805548", "exception": false, "start_time": "2022-05-12T12:19:18.683845", "status": "completed"}, "tags": []}, "outputs": [{"data": {"application/vnd.jupyter.widget-view+json": {"model_id": "a05e28af0506439f8843dceab189d0a8", "version_major": 2, "version_minor": 0}, "text/plain": ["  0%|          | 0/8 [00:00<?, ?it/s]"]}, "metadata": {}, "output_type": "display_data"}, {"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"599.393438pt\" height=\"215.984063pt\" viewBox=\"0 0 599.393438 215.984063\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:19.238784</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 215.984063 \n", "L 599.393438 215.984063 \n", "L 599.393438 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g id=\"patch_2\">\n", "    <path d=\"M 34.193438 188.638125 \n", "L 592.193437 188.638125 \n", "L 592.193437 22.318125 \n", "L 34.193438 22.318125 \n", "z\n", "\" style=\"fill: #eaeaf2\"/>\n", "   </g>\n", "   <g id=\"matplotlib.axis_1\">\n", "    <g id=\"xtick_1\">\n", "     <g id=\"line2d_1\">\n", "      <path d=\"M 105.551498 188.638125 \n", "L 105.551498 22.318125 \n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_1\">\n", "      <!-- \u221210 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(93.94392 206.496406)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-2212\" d=\"M 678 2272 \n", "L 4684 2272 \n", "L 4684 1741 \n", "L 678 1741 \n", "L 678 2272 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-31\" d=\"M 794 531 \n", "L 1825 531 \n", "L 1825 4091 \n", "L 703 3866 \n", "L 703 4441 \n", "L 1819 4666 \n", "L 2450 4666 \n", "L 2450 531 \n", "L 3481 531 \n", "L 3481 0 \n", "L 794 0 \n", "L 794 531 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-30\" d=\"M 2034 4250 \n", "Q 1547 4250 1301 3770 \n", "Q 1056 3291 1056 2328 \n", "Q 1056 1369 1301 889 \n", "Q 1547 409 2034 409 \n", "Q 2525 409 2770 889 \n", "Q 3016 1369 3016 2328 \n", "Q 3016 3291 2770 3770 \n", "Q 2525 4250 2034 4250 \n", "z\n", "M 2034 4750 \n", "Q 2819 4750 3233 4129 \n", "Q 3647 3509 3647 2328 \n", "Q 3647 1150 3233 529 \n", "Q 2819 -91 2034 -91 \n", "Q 1250 -91 836 529 \n", "Q 422 1150 422 2328 \n", "Q 422 3509 836 4129 \n", "Q 1250 4750 2034 4750 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"83.789062\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"147.412109\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_2\">\n", "     <g id=\"line2d_2\">\n", "      <path d=\"M 209.79675 188.638125 \n", "L 209.79675 22.318125 \n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_2\">\n", "      <!-- \u22125 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(201.688547 206.496406)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-35\" d=\"M 691 4666 \n", "L 3169 4666 \n", "L 3169 4134 \n", "L 1269 4134 \n", "L 1269 2991 \n", "Q 1406 3038 1543 3061 \n", "Q 1681 3084 1819 3084 \n", "Q 2600 3084 3056 2656 \n", "Q 3513 2228 3513 1497 \n", "Q 3513 744 3044 326 \n", "Q 2575 -91 1722 -91 \n", "Q 1428 -91 1123 -41 \n", "Q 819 9 494 109 \n", "L 494 744 \n", "Q 775 591 1075 516 \n", "Q 1375 441 1709 441 \n", "Q 2250 441 2565 725 \n", "Q 2881 1009 2881 1497 \n", "Q 2881 1984 2565 2268 \n", "Q 2250 2553 1709 2553 \n", "Q 1456 2553 1204 2497 \n", "Q 953 2441 691 2322 \n", "L 691 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"83.789062\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_3\">\n", "     <g id=\"line2d_3\">\n", "      <path d=\"M 314.042003 188.638125 \n", "L 314.042003 22.318125 \n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_3\">\n", "      <!-- 0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(310.542628 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_4\">\n", "     <g id=\"line2d_4\">\n", "      <path d=\"M 418.287255 188.638125 \n", "L 418.287255 22.318125 \n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_4\">\n", "      <!-- 5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(414.78788 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-35\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_5\">\n", "     <g id=\"line2d_5\">\n", "      <path d=\"M 522.532508 188.638125 \n", "L 522.532508 22.318125 \n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_5\">\n", "      <!-- 10 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(515.533758 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-31\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"63.623047\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"matplotlib.axis_2\">\n", "    <g id=\"ytick_1\">\n", "     <g id=\"line2d_6\">\n", "      <path d=\"M 34.193438 188.638125 \n", "L 592.193437 188.638125 \n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_6\">\n", "      <!-- 0.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(7.2 192.817266)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-2e\" d=\"M 684 794 \n", "L 1344 794 \n", "L 1344 0 \n", "L 684 0 \n", "L 684 794 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_2\">\n", "     <g id=\"line2d_7\">\n", "      <path d=\"M 34.193438 147.103553 \n", "L 592.193437 147.103553 \n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_7\">\n", "      <!-- 0.2 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(7.2 151.282693)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-32\" d=\"M 1228 531 \n", "L 3431 531 \n", "L 3431 0 \n", "L 469 0 \n", "L 469 531 \n", "Q 828 903 1448 1529 \n", "Q 2069 2156 2228 2338 \n", "Q 2531 2678 2651 2914 \n", "Q 2772 3150 2772 3378 \n", "Q 2772 3750 2511 3984 \n", "Q 2250 4219 1831 4219 \n", "Q 1534 4219 1204 4116 \n", "Q 875 4013 500 3803 \n", "L 500 4441 \n", "Q 881 4594 1212 4672 \n", "Q 1544 4750 1819 4750 \n", "Q 2544 4750 2975 4387 \n", "Q 3406 4025 3406 3419 \n", "Q 3406 3131 3298 2873 \n", "Q 3191 2616 2906 2266 \n", "Q 2828 2175 2409 1742 \n", "Q 1991 1309 1228 531 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-32\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_3\">\n", "     <g id=\"line2d_8\">\n", "      <path d=\"M 34.193438 105.56898 \n", "L 592.193437 105.56898 \n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_8\">\n", "      <!-- 0.4 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(7.2 109.748121)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-34\" d=\"M 2419 4116 \n", "L 825 1625 \n", "L 2419 1625 \n", "L 2419 4116 \n", "z\n", "M 2253 4666 \n", "L 3047 4666 \n", "L 3047 1625 \n", "L 3713 1625 \n", "L 3713 1100 \n", "L 3047 1100 \n", "L 3047 0 \n", "L 2419 0 \n", "L 2419 1100 \n", "L 313 1100 \n", "L 313 1709 \n", "L 2253 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-34\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_4\">\n", "     <g id=\"line2d_9\">\n", "      <path d=\"M 34.193438 64.034408 \n", "L 592.193437 64.034408 \n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_9\">\n", "      <!-- 0.6 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(7.2 68.213548)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-36\" d=\"M 2113 2584 \n", "Q 1688 2584 1439 2293 \n", "Q 1191 2003 1191 1497 \n", "Q 1191 994 1439 701 \n", "Q 1688 409 2113 409 \n", "Q 2538 409 2786 701 \n", "Q 3034 994 3034 1497 \n", "Q 3034 2003 2786 2293 \n", "Q 2538 2584 2113 2584 \n", "z\n", "M 3366 4563 \n", "L 3366 3988 \n", "Q 3128 4100 2886 4159 \n", "Q 2644 4219 2406 4219 \n", "Q 1781 4219 1451 3797 \n", "Q 1122 3375 1075 2522 \n", "Q 1259 2794 1537 2939 \n", "Q 1816 3084 2150 3084 \n", "Q 2853 3084 3261 2657 \n", "Q 3669 2231 3669 1497 \n", "Q 3669 778 3244 343 \n", "Q 2819 -91 2113 -91 \n", "Q 1303 -91 875 529 \n", "Q 447 1150 447 2328 \n", "Q 447 3434 972 4092 \n", "Q 1497 4750 2381 4750 \n", "Q 2619 4750 2861 4703 \n", "Q 3103 4656 3366 4563 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-36\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_5\">\n", "     <g id=\"line2d_10\">\n", "      <path d=\"M 34.193438 22.499835 \n", "L 592.193437 22.499835 \n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_10\">\n", "      <!-- 0.8 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(7.2 26.678976)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-38\" d=\"M 2034 2216 \n", "Q 1584 2216 1326 1975 \n", "Q 1069 1734 1069 1313 \n", "Q 1069 891 1326 650 \n", "Q 1584 409 2034 409 \n", "Q 2484 409 2743 651 \n", "Q 3003 894 3003 1313 \n", "Q 3003 1734 2745 1975 \n", "Q 2488 2216 2034 2216 \n", "z\n", "M 1403 2484 \n", "Q 997 2584 770 2862 \n", "Q 544 3141 544 3541 \n", "Q 544 4100 942 4425 \n", "Q 1341 4750 2034 4750 \n", "Q 2731 4750 3128 4425 \n", "Q 3525 4100 3525 3541 \n", "Q 3525 3141 3298 2862 \n", "Q 3072 2584 2669 2484 \n", "Q 3125 2378 3379 2068 \n", "Q 3634 1759 3634 1313 \n", "Q 3634 634 3220 271 \n", "Q 2806 -91 2034 -91 \n", "Q 1263 -91 848 271 \n", "Q 434 634 434 1313 \n", "Q 434 1759 690 2068 \n", "Q 947 2378 1403 2484 \n", "z\n", "M 1172 3481 \n", "Q 1172 3119 1398 2916 \n", "Q 1625 2713 2034 2713 \n", "Q 2441 2713 2670 2916 \n", "Q 2900 3119 2900 3481 \n", "Q 2900 3844 2670 4047 \n", "Q 2441 4250 2034 4250 \n", "Q 1625 4250 1398 4047 \n", "Q 1172 3844 1172 3481 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-38\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"patch_3\">\n", "    <path d=\"M 59.557074 188.638125 \n", "L 61.538614 188.638125 \n", "L 61.538614 188.432633 \n", "L 59.557074 188.432633 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_4\">\n", "    <path d=\"M 61.538614 188.638125 \n", "L 63.520134 188.638125 \n", "L 63.520134 188.378196 \n", "L 61.538614 188.378196 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_5\">\n", "    <path d=\"M 63.520115 188.638125 \n", "L 65.501655 188.638125 \n", "L 65.501655 188.365951 \n", "L 63.520115 188.365951 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_6\">\n", "    <path d=\"M 65.501675 188.638125 \n", "L 67.483215 188.638125 \n", "L 67.483215 188.34962 \n", "L 65.501675 188.34962 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_7\">\n", "    <path d=\"M 67.483215 188.638125 \n", "L 69.464735 188.638125 \n", "L 69.464735 188.299265 \n", "L 67.483215 188.299265 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_8\">\n", "    <path d=\"M 69.464715 188.638125 \n", "L 71.446256 188.638125 \n", "L 71.446256 188.281577 \n", "L 69.464715 188.281577 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_9\">\n", "    <path d=\"M 71.446275 188.638125 \n", "L 73.427816 188.638125 \n", "L 73.427816 188.272051 \n", "L 71.446275 188.272051 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_10\">\n", "    <path d=\"M 73.427796 188.638125 \n", "L 75.409336 188.638125 \n", "L 75.409336 188.229864 \n", "L 73.427796 188.229864 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_11\">\n", "    <path d=\"M 75.409356 188.638125 \n", "L 77.390876 188.638125 \n", "L 77.390876 188.135958 \n", "L 75.409356 188.135958 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_12\">\n", "    <path d=\"M 77.390876 188.638125 \n", "L 79.372416 188.638125 \n", "L 79.372416 188.055672 \n", "L 77.390876 188.055672 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_13\">\n", "    <path d=\"M 79.372397 188.638125 \n", "L 81.353937 188.638125 \n", "L 81.353937 188.078807 \n", "L 79.372397 188.078807 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_14\">\n", "    <path d=\"M 81.353957 188.638125 \n", "L 83.335477 188.638125 \n", "L 83.335477 188.005313 \n", "L 81.353957 188.005313 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_15\">\n", "    <path d=\"M 83.335477 188.638125 \n", "L 85.317017 188.638125 \n", "L 85.317017 187.862428 \n", "L 83.335477 187.862428 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_16\">\n", "    <path d=\"M 85.316997 188.638125 \n", "L 87.298538 188.638125 \n", "L 87.298538 187.821602 \n", "L 85.316997 187.821602 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_17\">\n", "    <path d=\"M 87.298558 188.638125 \n", "L 89.280078 188.638125 \n", "L 89.280078 187.858338 \n", "L 87.298558 187.858338 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_18\">\n", "    <path d=\"M 89.280078 188.638125 \n", "L 91.261618 188.638125 \n", "L 91.261618 187.714093 \n", "L 89.280078 187.714093 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_19\">\n", "    <path d=\"M 91.261598 188.638125 \n", "L 93.243138 188.638125 \n", "L 93.243138 187.628358 \n", "L 91.261598 187.628358 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_20\">\n", "    <path d=\"M 93.243158 188.638125 \n", "L 95.224679 188.638125 \n", "L 95.224679 187.531725 \n", "L 93.243158 187.531725 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_21\">\n", "    <path d=\"M 95.224679 188.638125 \n", "L 97.206219 188.638125 \n", "L 97.206219 187.401093 \n", "L 95.224679 187.401093 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_22\">\n", "    <path d=\"M 97.206199 188.638125 \n", "L 99.187739 188.638125 \n", "L 99.187739 187.284058 \n", "L 97.206199 187.284058 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_23\">\n", "    <path d=\"M 99.187759 188.638125 \n", "L 101.169299 188.638125 \n", "L 101.169299 187.048627 \n", "L 99.187759 187.048627 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_24\">\n", "    <path d=\"M 101.169299 188.638125 \n", "L 103.15082 188.638125 \n", "L 103.15082 186.879861 \n", "L 101.169299 186.879861 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_25\">\n", "    <path d=\"M 103.1508 188.638125 \n", "L 105.13234 188.638125 \n", "L 105.13234 186.728822 \n", "L 103.1508 186.728822 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_26\">\n", "    <path d=\"M 105.13236 188.638125 \n", "L 107.1139 188.638125 \n", "L 107.1139 186.585931 \n", "L 105.13236 186.585931 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_27\">\n", "    <path d=\"M 107.1139 188.638125 \n", "L 109.09542 188.638125 \n", "L 109.09542 186.436213 \n", "L 107.1139 186.436213 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_28\">\n", "    <path d=\"M 109.095401 188.638125 \n", "L 111.076941 188.638125 \n", "L 111.076941 186.199443 \n", "L 109.095401 186.199443 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_29\">\n", "    <path d=\"M 111.076961 188.638125 \n", "L 113.058501 188.638125 \n", "L 113.058501 185.965373 \n", "L 111.076961 185.965373 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_30\">\n", "    <path d=\"M 113.058501 188.638125 \n", "L 115.040021 188.638125 \n", "L 115.040021 185.732635 \n", "L 113.058501 185.732635 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_31\">\n", "    <path d=\"M 115.040001 188.638125 \n", "L 117.021542 188.638125 \n", "L 117.021542 185.497234 \n", "L 115.040001 185.497234 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_32\">\n", "    <path d=\"M 117.021561 188.638125 \n", "L 119.003102 188.638125 \n", "L 119.003102 185.173346 \n", "L 117.021561 185.173346 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_33\">\n", "    <path d=\"M 119.003102 188.638125 \n", "L 120.984622 188.638125 \n", "L 120.984622 184.680671 \n", "L 119.003102 184.680671 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_34\">\n", "    <path d=\"M 120.984602 188.638125 \n", "L 122.966142 188.638125 \n", "L 122.966142 184.262923 \n", "L 120.984602 184.262923 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_35\">\n", "    <path d=\"M 122.966162 188.638125 \n", "L 124.947703 188.638125 \n", "L 124.947703 183.919983 \n", "L 122.966162 183.919983 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_36\">\n", "    <path d=\"M 124.947683 188.638125 \n", "L 126.929223 188.638125 \n", "L 126.929223 183.494031 \n", "L 124.947683 183.494031 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_37\">\n", "    <path d=\"M 126.929243 188.638125 \n", "L 128.910763 188.638125 \n", "L 128.910763 182.971399 \n", "L 126.929243 182.971399 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_38\">\n", "    <path d=\"M 128.910763 188.638125 \n", "L 130.892303 188.638125 \n", "L 130.892303 182.299186 \n", "L 128.910763 182.299186 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_39\">\n", "    <path d=\"M 130.892283 188.638125 \n", "L 132.873824 188.638125 \n", "L 132.873824 181.780693 \n", "L 130.892283 181.780693 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_40\">\n", "    <path d=\"M 132.873844 188.638125 \n", "L 134.855364 188.638125 \n", "L 134.855364 180.991311 \n", "L 132.873844 180.991311 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_41\">\n", "    <path d=\"M 134.855364 188.638125 \n", "L 136.836904 188.638125 \n", "L 136.836904 180.339531 \n", "L 134.855364 180.339531 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_42\">\n", "    <path d=\"M 136.836884 188.638125 \n", "L 138.818424 188.638125 \n", "L 138.818424 179.781573 \n", "L 136.836884 179.781573 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_43\">\n", "    <path d=\"M 138.818444 188.638125 \n", "L 140.799965 188.638125 \n", "L 140.799965 178.473674 \n", "L 138.818444 178.473674 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_44\">\n", "    <path d=\"M 140.799965 188.638125 \n", "L 142.781505 188.638125 \n", "L 142.781505 177.488505 \n", "L 140.799965 177.488505 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_45\">\n", "    <path d=\"M 142.781485 188.638125 \n", "L 144.763025 188.638125 \n", "L 144.763025 176.281412 \n", "L 142.781485 176.281412 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_46\">\n", "    <path d=\"M 144.763045 188.638125 \n", "L 146.744565 188.638125 \n", "L 146.744565 175.11637 \n", "L 144.763045 175.11637 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_47\">\n", "    <path d=\"M 146.744565 188.638125 \n", "L 148.726106 188.638125 \n", "L 148.726106 174.142122 \n", "L 146.744565 174.142122 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_48\">\n", "    <path d=\"M 148.726106 188.638125 \n", "L 150.707646 188.638125 \n", "L 150.707646 172.564872 \n", "L 148.726106 172.564872 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_49\">\n", "    <path d=\"M 150.707646 188.638125 \n", "L 152.689166 188.638125 \n", "L 152.689166 170.999781 \n", "L 150.707646 170.999781 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_50\">\n", "    <path d=\"M 152.689186 188.638125 \n", "L 154.670706 188.638125 \n", "L 154.670706 169.165317 \n", "L 152.689186 169.165317 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_51\">\n", "    <path d=\"M 154.670706 188.638125 \n", "L 156.652247 188.638125 \n", "L 156.652247 166.982577 \n", "L 154.670706 166.982577 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_52\">\n", "    <path d=\"M 156.652247 188.638125 \n", "L 158.633767 188.638125 \n", "L 158.633767 164.696192 \n", "L 156.652247 164.696192 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_53\">\n", "    <path d=\"M 158.633777 188.638125 \n", "L 160.615317 188.638125 \n", "L 160.615317 162.355614 \n", "L 158.633777 162.355614 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_54\">\n", "    <path d=\"M 160.615327 188.638125 \n", "L 162.596848 188.638125 \n", "L 162.596848 159.76573 \n", "L 160.615327 159.76573 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_55\">\n", "    <path d=\"M 162.596848 188.638125 \n", "L 164.578368 188.638125 \n", "L 164.578368 157.004508 \n", "L 162.596848 157.004508 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_56\">\n", "    <path d=\"M 164.578378 188.638125 \n", "L 166.559918 188.638125 \n", "L 166.559918 153.71544 \n", "L 164.578378 153.71544 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_57\">\n", "    <path d=\"M 166.559928 188.638125 \n", "L 168.541448 188.638125 \n", "L 168.541448 150.442353 \n", "L 166.559928 150.442353 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_58\">\n", "    <path d=\"M 168.541448 188.638125 \n", "L 170.522969 188.638125 \n", "L 170.522969 146.513497 \n", "L 168.541448 146.513497 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_59\">\n", "    <path d=\"M 170.522979 188.638125 \n", "L 172.504519 188.638125 \n", "L 172.504519 142.28412 \n", "L 170.522979 142.28412 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_60\">\n", "    <path d=\"M 172.504529 188.638125 \n", "L 174.486049 188.638125 \n", "L 174.486049 138.093743 \n", "L 172.504529 138.093743 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_61\">\n", "    <path d=\"M 174.486049 188.638125 \n", "L 176.467589 188.638125 \n", "L 176.467589 132.929489 \n", "L 174.486049 132.929489 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_62\">\n", "    <path d=\"M 176.467589 188.638125 \n", "L 178.44911 188.638125 \n", "L 178.44911 127.065185 \n", "L 176.467589 127.065185 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_63\">\n", "    <path d=\"M 178.44913 188.638125 \n", "L 180.43065 188.638125 \n", "L 180.43065 121.689715 \n", "L 178.44913 121.689715 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_64\">\n", "    <path d=\"M 180.43065 188.638125 \n", "L 182.41219 188.638125 \n", "L 182.41219 114.907468 \n", "L 180.43065 114.907468 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_65\">\n", "    <path d=\"M 182.41219 188.638125 \n", "L 184.39371 188.638125 \n", "L 184.39371 107.067078 \n", "L 182.41219 107.067078 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_66\">\n", "    <path d=\"M 184.39373 188.638125 \n", "L 186.375251 188.638125 \n", "L 186.375251 99.315515 \n", "L 184.39373 99.315515 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_67\">\n", "    <path d=\"M 186.375251 188.638125 \n", "L 188.356791 188.638125 \n", "L 188.356791 90.085171 \n", "L 186.375251 90.085171 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_68\">\n", "    <path d=\"M 188.356791 188.638125 \n", "L 190.338311 188.638125 \n", "L 190.338311 80.444215 \n", "L 188.356791 80.444215 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_69\">\n", "    <path d=\"M 190.338321 188.638125 \n", "L 192.319861 188.638125 \n", "L 192.319861 69.570033 \n", "L 190.338321 69.570033 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_70\">\n", "    <path d=\"M 192.319871 188.638125 \n", "L 194.301392 188.638125 \n", "L 194.301392 57.633172 \n", "L 192.319871 57.633172 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_71\">\n", "    <path d=\"M 194.301392 188.638125 \n", "L 196.282912 188.638125 \n", "L 196.282912 45.540407 \n", "L 194.301392 45.540407 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_72\">\n", "    <path d=\"M 196.282922 188.638125 \n", "L 198.264462 188.638125 \n", "L 198.264462 30.238125 \n", "L 196.282922 30.238125 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_73\">\n", "    <path d=\"M 198.264472 188.638125 \n", "L 200.245993 188.638125 \n", "L 200.245993 165.192913 \n", "L 198.264472 165.192913 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_74\">\n", "    <path d=\"M 200.245993 188.638125 \n", "L 202.227533 188.638125 \n", "L 202.227533 188.544225 \n", "L 200.245993 188.544225 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_75\">\n", "    <path d=\"M 202.227533 188.638125 \n", "L 204.209053 188.638125 \n", "L 204.209053 188.561916 \n", "L 202.227533 188.561916 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_76\">\n", "    <path d=\"M 204.209073 188.638125 \n", "L 206.190593 188.638125 \n", "L 206.190593 188.530616 \n", "L 204.209073 188.530616 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_77\">\n", "    <path d=\"M 206.190593 188.638125 \n", "L 208.172134 188.638125 \n", "L 208.172134 188.53606 \n", "L 206.190593 188.53606 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_78\">\n", "    <path d=\"M 208.172134 188.638125 \n", "L 210.153654 188.638125 \n", "L 210.153654 188.533337 \n", "L 208.172134 188.533337 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_79\">\n", "    <path d=\"M 210.153674 188.638125 \n", "L 212.135194 188.638125 \n", "L 212.135194 188.481624 \n", "L 210.153674 188.481624 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_80\">\n", "    <path d=\"M 212.135194 188.638125 \n", "L 214.116734 188.638125 \n", "L 214.116734 188.39589 \n", "L 212.135194 188.39589 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_81\">\n", "    <path d=\"M 214.116734 188.638125 \n", "L 216.098255 188.638125 \n", "L 216.098255 188.276131 \n", "L 214.116734 188.276131 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_82\">\n", "    <path d=\"M 216.098265 188.638125 \n", "L 218.079805 188.638125 \n", "L 218.079805 188.223059 \n", "L 216.098265 188.223059 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_83\">\n", "    <path d=\"M 218.079815 188.638125 \n", "L 220.061335 188.638125 \n", "L 220.061335 188.242109 \n", "L 218.079815 188.242109 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_84\">\n", "    <path d=\"M 220.061335 188.638125 \n", "L 222.042855 188.638125 \n", "L 222.042855 188.179509 \n", "L 220.061335 188.179509 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_85\">\n", "    <path d=\"M 222.042865 188.638125 \n", "L 224.024406 188.638125 \n", "L 224.024406 188.220337 \n", "L 222.042865 188.220337 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_86\">\n", "    <path d=\"M 224.024416 188.638125 \n", "L 226.005936 188.638125 \n", "L 226.005936 188.179509 \n", "L 224.024416 188.179509 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_87\">\n", "    <path d=\"M 226.005936 188.638125 \n", "L 227.987476 188.638125 \n", "L 227.987476 188.14549 \n", "L 226.005936 188.14549 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_88\">\n", "    <path d=\"M 227.987476 188.638125 \n", "L 229.968997 188.638125 \n", "L 229.968997 187.949521 \n", "L 227.987476 187.949521 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_89\">\n", "    <path d=\"M 229.969006 188.638125 \n", "L 231.950547 188.638125 \n", "L 231.950547 187.904613 \n", "L 229.969006 187.904613 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_90\">\n", "    <path d=\"M 231.950547 188.638125 \n", "L 233.932067 188.638125 \n", "L 233.932067 187.873311 \n", "L 231.950547 187.873311 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_91\">\n", "    <path d=\"M 233.932067 188.638125 \n", "L 235.913607 188.638125 \n", "L 235.913607 187.933192 \n", "L 233.932067 187.933192 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_92\">\n", "    <path d=\"M 235.913607 188.638125 \n", "L 237.895147 188.638125 \n", "L 237.895147 187.501794 \n", "L 235.913607 187.501794 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_93\">\n", "    <path d=\"M 237.895138 188.638125 \n", "L 239.876678 188.638125 \n", "L 239.876678 187.425585 \n", "L 237.895138 187.425585 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_94\">\n", "    <path d=\"M 239.876683 188.638125 \n", "L 241.858203 188.638125 \n", "L 241.858203 187.537174 \n", "L 239.876683 187.537174 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_95\">\n", "    <path d=\"M 241.858208 188.638125 \n", "L 243.839748 188.638125 \n", "L 243.839748 187.420142 \n", "L 241.858208 187.420142 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_96\">\n", "    <path d=\"M 243.839738 188.638125 \n", "L 245.821279 188.638125 \n", "L 245.821279 187.289498 \n", "L 243.839738 187.289498 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_97\">\n", "    <path d=\"M 245.821279 188.638125 \n", "L 247.802819 188.638125 \n", "L 247.802819 186.840409 \n", "L 245.821279 186.840409 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_98\">\n", "    <path d=\"M 247.802809 188.638125 \n", "L 249.784349 188.638125 \n", "L 249.784349 186.885318 \n", "L 247.802809 186.885318 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_99\">\n", "    <path d=\"M 249.784354 188.638125 \n", "L 251.765874 188.638125 \n", "L 251.765874 186.867622 \n", "L 249.784354 186.867622 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_100\">\n", "    <path d=\"M 251.765879 188.638125 \n", "L 253.74742 188.638125 \n", "L 253.74742 186.811831 \n", "L 251.765879 186.811831 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_101\">\n", "    <path d=\"M 253.74741 188.638125 \n", "L 255.72895 188.638125 \n", "L 255.72895 186.713848 \n", "L 253.74741 186.713848 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_102\">\n", "    <path d=\"M 255.72895 188.638125 \n", "L 257.71049 188.638125 \n", "L 257.71049 186.798222 \n", "L 255.72895 186.798222 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_103\">\n", "    <path d=\"M 257.71049 188.638125 \n", "L 259.69201 188.638125 \n", "L 259.69201 186.7397 \n", "L 257.71049 186.7397 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_104\">\n", "    <path d=\"M 259.69201 188.638125 \n", "L 261.673551 188.638125 \n", "L 261.673551 186.425342 \n", "L 259.69201 186.425342 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_105\">\n", "    <path d=\"M 261.673551 188.638125 \n", "L 263.655091 188.638125 \n", "L 263.655091 186.159972 \n", "L 261.673551 186.159972 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_106\">\n", "    <path d=\"M 263.655081 188.638125 \n", "L 265.636621 188.638125 \n", "L 265.636621 185.646922 \n", "L 263.655081 185.646922 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_107\">\n", "    <path d=\"M 265.636626 188.638125 \n", "L 267.618146 188.638125 \n", "L 267.618146 186.410367 \n", "L 265.636626 186.410367 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_108\">\n", "    <path d=\"M 267.618151 188.638125 \n", "L 269.599692 188.638125 \n", "L 269.599692 186.131393 \n", "L 267.618151 186.131393 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_109\">\n", "    <path d=\"M 269.599682 188.638125 \n", "L 271.581222 188.638125 \n", "L 271.581222 185.636035 \n", "L 269.599682 185.636035 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_110\">\n", "    <path d=\"M 271.581217 188.638125 \n", "L 273.562757 188.638125 \n", "L 273.562757 184.978728 \n", "L 271.581217 184.978728 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_111\">\n", "    <path d=\"M 273.562752 188.638125 \n", "L 275.544292 188.638125 \n", "L 275.544292 186.042937 \n", "L 273.562752 186.042937 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_112\">\n", "    <path d=\"M 275.544288 188.638125 \n", "L 277.525828 188.638125 \n", "L 277.525828 185.403321 \n", "L 275.544288 185.403321 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_113\">\n", "    <path d=\"M 277.52582 188.638125 \n", "L 279.507361 188.638125 \n", "L 279.507361 185.31759 \n", "L 277.52582 185.31759 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_114\">\n", "    <path d=\"M 279.507356 188.638125 \n", "L 281.488896 188.638125 \n", "L 281.488896 184.903884 \n", "L 279.507356 184.903884 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_115\">\n", "    <path d=\"M 281.488888 188.638125 \n", "L 283.470429 188.638125 \n", "L 283.470429 185.418291 \n", "L 281.488888 185.418291 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_116\">\n", "    <path d=\"M 283.470424 188.638125 \n", "L 285.451964 188.638125 \n", "L 285.451964 184.925658 \n", "L 283.470424 184.925658 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_117\">\n", "    <path d=\"M 285.451959 188.638125 \n", "L 287.433499 188.638125 \n", "L 287.433499 185.480891 \n", "L 285.451959 185.480891 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_118\">\n", "    <path d=\"M 287.433492 188.638125 \n", "L 289.415032 188.638125 \n", "L 289.415032 184.445269 \n", "L 287.433492 184.445269 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_119\">\n", "    <path d=\"M 289.415024 188.638125 \n", "L 291.396565 188.638125 \n", "L 291.396565 183.892748 \n", "L 289.415024 183.892748 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_120\">\n", "    <path d=\"M 291.39656 188.638125 \n", "L 293.3781 188.638125 \n", "L 293.3781 184.299653 \n", "L 291.39656 184.299653 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_121\">\n", "    <path d=\"M 293.378092 188.638125 \n", "L 295.359633 188.638125 \n", "L 295.359633 184.347284 \n", "L 293.378092 184.347284 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_122\">\n", "    <path d=\"M 295.359628 188.638125 \n", "L 297.341168 188.638125 \n", "L 297.341168 184.237056 \n", "L 295.359628 184.237056 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_123\">\n", "    <path d=\"M 297.341163 188.638125 \n", "L 299.322703 188.638125 \n", "L 299.322703 184.17037 \n", "L 297.341163 184.17037 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_124\">\n", "    <path d=\"M 299.322696 188.638125 \n", "L 301.304236 188.638125 \n", "L 301.304236 183.213674 \n", "L 299.322696 183.213674 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_125\">\n", "    <path d=\"M 301.304231 188.638125 \n", "L 303.285771 188.638125 \n", "L 303.285771 184.40444 \n", "L 301.304231 184.40444 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_126\">\n", "    <path d=\"M 303.285764 188.638125 \n", "L 305.267304 188.638125 \n", "L 305.267304 184.171731 \n", "L 303.285764 184.171731 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_127\">\n", "    <path d=\"M 305.267298 188.638125 \n", "L 307.248839 188.638125 \n", "L 307.248839 183.962156 \n", "L 305.267298 183.962156 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_128\">\n", "    <path d=\"M 307.248832 188.638125 \n", "L 309.230373 188.638125 \n", "L 309.230373 184.053335 \n", "L 307.248832 184.053335 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_129\">\n", "    <path d=\"M 309.230366 188.638125 \n", "L 311.211907 188.638125 \n", "L 311.211907 183.529397 \n", "L 309.230366 183.529397 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_130\">\n", "    <path d=\"M 311.2119 188.638125 \n", "L 313.193441 188.638125 \n", "L 313.193441 183.638267 \n", "L 311.2119 183.638267 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_131\">\n", "    <path d=\"M 313.193434 188.638125 \n", "L 315.174975 188.638125 \n", "L 315.174975 181.165556 \n", "L 313.193434 181.165556 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_132\">\n", "    <path d=\"M 315.174968 188.638125 \n", "L 317.156509 188.638125 \n", "L 317.156509 183.763468 \n", "L 315.174968 183.763468 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_133\">\n", "    <path d=\"M 317.156502 188.638125 \n", "L 319.138043 188.638125 \n", "L 319.138043 183.151075 \n", "L 317.156502 183.151075 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_134\">\n", "    <path d=\"M 319.138036 188.638125 \n", "L 321.119577 188.638125 \n", "L 321.119577 183.804293 \n", "L 319.138036 183.804293 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_135\">\n", "    <path d=\"M 321.119571 188.638125 \n", "L 323.101111 188.638125 \n", "L 323.101111 183.76347 \n", "L 321.119571 183.76347 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_136\">\n", "    <path d=\"M 323.101105 188.638125 \n", "L 325.082645 188.638125 \n", "L 325.082645 184.473844 \n", "L 323.101105 184.473844 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_137\">\n", "    <path d=\"M 325.082639 188.638125 \n", "L 327.064179 188.638125 \n", "L 327.064179 184.394914 \n", "L 325.082639 184.394914 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_138\">\n", "    <path d=\"M 327.064172 188.638125 \n", "L 329.045712 188.638125 \n", "L 329.045712 182.793164 \n", "L 327.064172 182.793164 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_139\">\n", "    <path d=\"M 329.045707 188.638125 \n", "L 331.027247 188.638125 \n", "L 331.027247 184.238416 \n", "L 329.045707 184.238416 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_140\">\n", "    <path d=\"M 331.027242 188.638125 \n", "L 333.008783 188.638125 \n", "L 333.008783 183.827429 \n", "L 331.027242 183.827429 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_141\">\n", "    <path d=\"M 333.008775 188.638125 \n", "L 334.990315 188.638125 \n", "L 334.990315 185.176057 \n", "L 333.008775 185.176057 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_142\">\n", "    <path d=\"M 334.99031 188.638125 \n", "L 336.971851 188.638125 \n", "L 336.971851 182.820377 \n", "L 334.99031 182.820377 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_143\">\n", "    <path d=\"M 336.971843 188.638125 \n", "L 338.953383 188.638125 \n", "L 338.953383 184.424856 \n", "L 336.971843 184.424856 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_144\">\n", "    <path d=\"M 338.953376 188.638125 \n", "L 340.934916 188.638125 \n", "L 340.934916 184.318703 \n", "L 338.953376 184.318703 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_145\">\n", "    <path d=\"M 340.934911 188.638125 \n", "L 342.916451 188.638125 \n", "L 342.916451 185.171977 \n", "L 340.934911 185.171977 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_146\">\n", "    <path d=\"M 342.916446 188.638125 \n", "L 344.897987 188.638125 \n", "L 344.897987 185.351608 \n", "L 342.916446 185.351608 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_147\">\n", "    <path d=\"M 344.897979 188.638125 \n", "L 346.879519 188.638125 \n", "L 346.879519 184.94471 \n", "L 344.897979 184.94471 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_148\">\n", "    <path d=\"M 346.879514 188.638125 \n", "L 348.861055 188.638125 \n", "L 348.861055 184.793653 \n", "L 346.879514 184.793653 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_149\">\n", "    <path d=\"M 348.861047 188.638125 \n", "L 350.842587 188.638125 \n", "L 350.842587 184.978728 \n", "L 348.861047 184.978728 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_150\">\n", "    <path d=\"M 350.842583 188.638125 \n", "L 352.824123 188.638125 \n", "L 352.824123 185.980336 \n", "L 350.842583 185.980336 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_151\">\n", "    <path d=\"M 352.824118 188.638125 \n", "L 354.805658 188.638125 \n", "L 354.805658 185.259068 \n", "L 352.824118 185.259068 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_152\">\n", "    <path d=\"M 354.805653 188.638125 \n", "L 356.787193 188.638125 \n", "L 356.787193 185.142037 \n", "L 354.805653 185.142037 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_153\">\n", "    <path d=\"M 356.787183 188.638125 \n", "L 358.768724 188.638125 \n", "L 358.768724 185.800701 \n", "L 356.787183 185.800701 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_154\">\n", "    <path d=\"M 358.768729 188.638125 \n", "L 360.750249 188.638125 \n", "L 360.750249 186.127304 \n", "L 358.768729 186.127304 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_155\">\n", "    <path d=\"M 360.750254 188.638125 \n", "L 362.731794 188.638125 \n", "L 362.731794 185.743544 \n", "L 360.750254 185.743544 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_156\">\n", "    <path d=\"M 362.731784 188.638125 \n", "L 364.713324 188.638125 \n", "L 364.713324 185.934067 \n", "L 362.731784 185.934067 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_157\">\n", "    <path d=\"M 364.713324 188.638125 \n", "L 366.694865 188.638125 \n", "L 366.694865 186.068793 \n", "L 364.713324 186.068793 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_158\">\n", "    <path d=\"M 366.694865 188.638125 \n", "L 368.676385 188.638125 \n", "L 368.676385 186.511072 \n", "L 366.694865 186.511072 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_159\">\n", "    <path d=\"M 368.676385 188.638125 \n", "L 370.657925 188.638125 \n", "L 370.657925 186.682548 \n", "L 368.676385 186.682548 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_160\">\n", "    <path d=\"M 370.657925 188.638125 \n", "L 372.639465 188.638125 \n", "L 372.639465 186.626752 \n", "L 370.657925 186.626752 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_161\">\n", "    <path d=\"M 372.639455 188.638125 \n", "L 374.620996 188.638125 \n", "L 374.620996 186.651248 \n", "L 372.639455 186.651248 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_162\">\n", "    <path d=\"M 374.621001 188.638125 \n", "L 376.602521 188.638125 \n", "L 376.602521 186.779165 \n", "L 374.621001 186.779165 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_163\">\n", "    <path d=\"M 376.602526 188.638125 \n", "L 378.584066 188.638125 \n", "L 378.584066 187.04454 \n", "L 376.602526 187.04454 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_164\">\n", "    <path d=\"M 378.584056 188.638125 \n", "L 380.565596 188.638125 \n", "L 380.565596 186.947918 \n", "L 378.584056 186.947918 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_165\">\n", "    <path d=\"M 380.565596 188.638125 \n", "L 382.547137 188.638125 \n", "L 382.547137 187.377955 \n", "L 380.565596 187.377955 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_166\">\n", "    <path d=\"M 382.547127 188.638125 \n", "L 384.528667 188.638125 \n", "L 384.528667 187.315355 \n", "L 382.547127 187.315355 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_167\">\n", "    <path d=\"M 384.528672 188.638125 \n", "L 386.510192 188.638125 \n", "L 386.510192 187.342569 \n", "L 384.528672 187.342569 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_168\">\n", "    <path d=\"M 386.510197 188.638125 \n", "L 388.491737 188.638125 \n", "L 388.491737 187.552147 \n", "L 386.510197 187.552147 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_169\">\n", "    <path d=\"M 388.491728 188.638125 \n", "L 390.473268 188.638125 \n", "L 390.473268 187.531734 \n", "L 388.491728 187.531734 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_170\">\n", "    <path d=\"M 390.473268 188.638125 \n", "L 392.454808 188.638125 \n", "L 392.454808 187.614747 \n", "L 390.473268 187.614747 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_171\">\n", "    <path d=\"M 392.454808 188.638125 \n", "L 394.436328 188.638125 \n", "L 394.436328 187.705923 \n", "L 392.454808 187.705923 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_172\">\n", "    <path d=\"M 394.436328 188.638125 \n", "L 396.417869 188.638125 \n", "L 396.417869 187.228259 \n", "L 394.436328 187.228259 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_173\">\n", "    <path d=\"M 396.417874 188.638125 \n", "L 398.399399 188.638125 \n", "L 398.399399 186.501546 \n", "L 396.417874 186.501546 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_174\">\n", "    <path d=\"M 398.399399 188.638125 \n", "L 400.380939 188.638125 \n", "L 400.380939 186.63084 \n", "L 398.399399 186.63084 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_175\">\n", "    <path d=\"M 400.380939 188.638125 \n", "L 402.362459 188.638125 \n", "L 402.362459 176.028226 \n", "L 400.380939 176.028226 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_176\">\n", "    <path d=\"M 402.362469 188.638125 \n", "L 404.34401 188.638125 \n", "L 404.34401 177.187752 \n", "L 402.362469 177.187752 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_177\">\n", "    <path d=\"M 404.34402 188.638125 \n", "L 406.32554 188.638125 \n", "L 406.32554 177.996056 \n", "L 404.34402 177.996056 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_178\">\n", "    <path d=\"M 406.32554 188.638125 \n", "L 408.30706 188.638125 \n", "L 408.30706 173.224817 \n", "L 406.32554 173.224817 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_179\">\n", "    <path d=\"M 408.30707 188.638125 \n", "L 410.28861 188.638125 \n", "L 410.28861 174.196557 \n", "L 408.30707 174.196557 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_180\">\n", "    <path d=\"M 410.28862 188.638125 \n", "L 412.270141 188.638125 \n", "L 412.270141 175.585941 \n", "L 410.28862 175.585941 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_181\">\n", "    <path d=\"M 412.270141 188.638125 \n", "L 414.251681 188.638125 \n", "L 414.251681 176.706004 \n", "L 412.270141 176.706004 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_182\">\n", "    <path d=\"M 414.251681 188.638125 \n", "L 416.233201 188.638125 \n", "L 416.233201 179.542014 \n", "L 414.251681 179.542014 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_183\">\n", "    <path d=\"M 416.233221 188.638125 \n", "L 418.214741 188.638125 \n", "L 418.214741 181.298908 \n", "L 416.233221 181.298908 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_184\">\n", "    <path d=\"M 418.214741 188.638125 \n", "L 420.196282 188.638125 \n", "L 420.196282 182.020207 \n", "L 418.214741 182.020207 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_185\">\n", "    <path d=\"M 420.196282 188.638125 \n", "L 422.177802 188.638125 \n", "L 422.177802 182.520977 \n", "L 420.196282 182.520977 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_186\">\n", "    <path d=\"M 422.177822 188.638125 \n", "L 424.159342 188.638125 \n", "L 424.159342 183.127929 \n", "L 422.177822 183.127929 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_187\">\n", "    <path d=\"M 424.159342 188.638125 \n", "L 426.140882 188.638125 \n", "L 426.140882 183.699522 \n", "L 424.159342 183.699522 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_188\">\n", "    <path d=\"M 426.140882 188.638125 \n", "L 428.122403 188.638125 \n", "L 428.122403 184.035634 \n", "L 426.140882 184.035634 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_189\">\n", "    <path d=\"M 428.122413 188.638125 \n", "L 430.103953 188.638125 \n", "L 430.103953 185.206007 \n", "L 428.122413 185.206007 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_190\">\n", "    <path d=\"M 430.103963 188.638125 \n", "L 432.085483 188.638125 \n", "L 432.085483 187.337125 \n", "L 430.103963 187.337125 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_191\">\n", "    <path d=\"M 432.085483 188.638125 \n", "L 434.067004 188.638125 \n", "L 434.067004 187.481378 \n", "L 432.085483 187.481378 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_192\">\n", "    <path d=\"M 434.067014 188.638125 \n", "L 436.048554 188.638125 \n", "L 436.048554 187.55351 \n", "L 434.067014 187.55351 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_193\">\n", "    <path d=\"M 436.048564 188.638125 \n", "L 438.030084 188.638125 \n", "L 438.030084 187.565753 \n", "L 436.048564 187.565753 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_194\">\n", "    <path d=\"M 438.030084 188.638125 \n", "L 440.011624 188.638125 \n", "L 440.011624 187.673267 \n", "L 438.030084 187.673267 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_195\">\n", "    <path d=\"M 440.011624 188.638125 \n", "L 441.993145 188.638125 \n", "L 441.993145 187.82432 \n", "L 440.011624 187.82432 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_196\">\n", "    <path d=\"M 441.993165 188.638125 \n", "L 443.974685 188.638125 \n", "L 443.974685 187.865146 \n", "L 441.993165 187.865146 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_197\">\n", "    <path d=\"M 443.974685 188.638125 \n", "L 445.956225 188.638125 \n", "L 445.956225 187.988989 \n", "L 443.974685 187.988989 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_198\">\n", "    <path d=\"M 445.956225 188.638125 \n", "L 447.937745 188.638125 \n", "L 447.937745 187.99579 \n", "L 445.956225 187.99579 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_199\">\n", "    <path d=\"M 447.937765 188.638125 \n", "L 449.919286 188.638125 \n", "L 449.919286 188.018925 \n", "L 447.937765 188.018925 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_200\">\n", "    <path d=\"M 449.919286 188.638125 \n", "L 451.900826 188.638125 \n", "L 451.900826 188.047507 \n", "L 449.919286 188.047507 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_201\">\n", "    <path d=\"M 451.900826 188.638125 \n", "L 453.882346 188.638125 \n", "L 453.882346 188.213531 \n", "L 451.900826 188.213531 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_202\">\n", "    <path d=\"M 453.882356 188.638125 \n", "L 455.863896 188.638125 \n", "L 455.863896 188.164542 \n", "L 453.882356 188.164542 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_203\">\n", "    <path d=\"M 455.863906 188.638125 \n", "L 457.845427 188.638125 \n", "L 457.845427 188.204005 \n", "L 455.863906 188.204005 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_204\">\n", "    <path d=\"M 457.845427 188.638125 \n", "L 459.826947 188.638125 \n", "L 459.826947 188.293823 \n", "L 457.845427 188.293823 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_205\">\n", "    <path d=\"M 459.826957 188.638125 \n", "L 461.808497 188.638125 \n", "L 461.808497 188.310155 \n", "L 459.826957 188.310155 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_206\">\n", "    <path d=\"M 461.808507 188.638125 \n", "L 463.790027 188.638125 \n", "L 463.790027 188.329206 \n", "L 461.808507 188.329206 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_207\">\n", "    <path d=\"M 463.790027 188.638125 \n", "L 465.771548 188.638125 \n", "L 465.771548 188.379558 \n", "L 463.790027 188.379558 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_208\">\n", "    <path d=\"M 465.771558 188.638125 \n", "L 467.753098 188.638125 \n", "L 467.753098 188.39589 \n", "L 465.771558 188.39589 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_209\">\n", "    <path d=\"M 467.753108 188.638125 \n", "L 469.734628 188.638125 \n", "L 469.734628 188.42855 \n", "L 467.753108 188.42855 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_210\">\n", "    <path d=\"M 469.734628 188.638125 \n", "L 471.716169 188.638125 \n", "L 471.716169 188.429912 \n", "L 469.734628 188.429912 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_211\">\n", "    <path d=\"M 471.716169 188.638125 \n", "L 473.697689 188.638125 \n", "L 473.697689 188.446241 \n", "L 471.716169 188.446241 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_212\">\n", "    <path d=\"M 473.697709 188.638125 \n", "L 475.679229 188.638125 \n", "L 475.679229 188.462572 \n", "L 473.697709 188.462572 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_213\">\n", "    <path d=\"M 475.679229 188.638125 \n", "L 477.660769 188.638125 \n", "L 477.660769 188.496594 \n", "L 475.679229 188.496594 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_214\">\n", "    <path d=\"M 477.660769 188.638125 \n", "L 479.64229 188.638125 \n", "L 479.64229 188.484346 \n", "L 477.660769 188.484346 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_215\">\n", "    <path d=\"M 479.64231 188.638125 \n", "L 481.62383 188.638125 \n", "L 481.62383 188.489789 \n", "L 479.64231 188.489789 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_216\">\n", "    <path d=\"M 481.62381 188.638125 \n", "L 483.60535 188.638125 \n", "L 483.60535 188.507481 \n", "L 481.62381 188.507481 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_217\">\n", "    <path d=\"M 483.60537 188.638125 \n", "L 485.58691 188.638125 \n", "L 485.58691 188.544225 \n", "L 483.60537 188.544225 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_218\">\n", "    <path d=\"M 485.58691 188.638125 \n", "L 487.568431 188.638125 \n", "L 487.568431 188.548307 \n", "L 485.58691 188.548307 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_219\">\n", "    <path d=\"M 487.568411 188.638125 \n", "L 489.549951 188.638125 \n", "L 489.549951 188.55239 \n", "L 487.568411 188.55239 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_220\">\n", "    <path d=\"M 489.549971 188.638125 \n", "L 491.531511 188.638125 \n", "L 491.531511 188.56736 \n", "L 489.549971 188.56736 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_221\">\n", "    <path d=\"M 491.531511 188.638125 \n", "L 493.513031 188.638125 \n", "L 493.513031 188.565998 \n", "L 491.531511 188.565998 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_222\">\n", "    <path d=\"M 493.513012 188.638125 \n", "L 495.494552 188.638125 \n", "L 495.494552 188.572803 \n", "L 493.513012 188.572803 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_223\">\n", "    <path d=\"M 495.494572 188.638125 \n", "L 497.476112 188.638125 \n", "L 497.476112 188.578247 \n", "L 495.494572 188.578247 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_224\">\n", "    <path d=\"M 497.476112 188.638125 \n", "L 499.457632 188.638125 \n", "L 499.457632 188.591855 \n", "L 497.476112 188.591855 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_225\">\n", "    <path d=\"M 499.457612 188.638125 \n", "L 501.439153 188.638125 \n", "L 501.439153 188.601381 \n", "L 499.457612 188.601381 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_226\">\n", "    <path d=\"M 501.439172 188.638125 \n", "L 503.420713 188.638125 \n", "L 503.420713 188.593216 \n", "L 501.439172 188.593216 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_227\">\n", "    <path d=\"M 503.420693 188.638125 \n", "L 505.402233 188.638125 \n", "L 505.402233 188.604103 \n", "L 503.420693 188.604103 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_228\">\n", "    <path d=\"M 505.402253 188.638125 \n", "L 507.383773 188.638125 \n", "L 507.383773 188.606825 \n", "L 505.402253 188.606825 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_229\">\n", "    <path d=\"M 507.383773 188.638125 \n", "L 509.365314 188.638125 \n", "L 509.365314 188.601381 \n", "L 507.383773 188.601381 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_230\">\n", "    <path d=\"M 509.365294 188.638125 \n", "L 511.346834 188.638125 \n", "L 511.346834 188.608186 \n", "L 509.365294 188.608186 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_231\">\n", "    <path d=\"M 511.346854 188.638125 \n", "L 513.328374 188.638125 \n", "L 513.328374 188.616351 \n", "L 511.346854 188.616351 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_232\">\n", "    <path d=\"M 513.328374 188.638125 \n", "L 515.309914 188.638125 \n", "L 515.309914 188.606825 \n", "L 513.328374 188.606825 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_233\">\n", "    <path d=\"M 515.309894 188.638125 \n", "L 517.291435 188.638125 \n", "L 517.291435 188.620434 \n", "L 515.309894 188.620434 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_234\">\n", "    <path d=\"M 517.291455 188.638125 \n", "L 519.272975 188.638125 \n", "L 519.272975 188.621794 \n", "L 517.291455 188.621794 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_235\">\n", "    <path d=\"M 519.272975 188.638125 \n", "L 521.254515 188.638125 \n", "L 521.254515 188.623155 \n", "L 519.272975 188.623155 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_236\">\n", "    <path d=\"M 521.254495 188.638125 \n", "L 523.236035 188.638125 \n", "L 523.236035 188.631321 \n", "L 521.254495 188.631321 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_237\">\n", "    <path d=\"M 523.236055 188.638125 \n", "L 525.217576 188.638125 \n", "L 525.217576 188.624516 \n", "L 523.236055 188.624516 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_238\">\n", "    <path d=\"M 525.217576 188.638125 \n", "L 527.199116 188.638125 \n", "L 527.199116 188.628599 \n", "L 525.217576 188.628599 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_239\">\n", "    <path d=\"M 527.199096 188.638125 \n", "L 529.180636 188.638125 \n", "L 529.180636 188.621795 \n", "L 527.199096 188.621795 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_240\">\n", "    <path d=\"M 529.180656 188.638125 \n", "L 531.162196 188.638125 \n", "L 531.162196 188.624516 \n", "L 529.180656 188.624516 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_241\">\n", "    <path d=\"M 531.162196 188.638125 \n", "L 533.143717 188.638125 \n", "L 533.143717 188.635403 \n", "L 531.162196 188.635403 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_242\">\n", "    <path d=\"M 533.143697 188.638125 \n", "L 535.125237 188.638125 \n", "L 535.125237 188.627238 \n", "L 533.143697 188.627238 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_243\">\n", "    <path d=\"M 535.125257 188.638125 \n", "L 537.106797 188.638125 \n", "L 537.106797 188.636764 \n", "L 535.125257 188.636764 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_244\">\n", "    <path d=\"M 537.106797 188.638125 \n", "L 539.088317 188.638125 \n", "L 539.088317 188.634042 \n", "L 537.106797 188.634042 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_245\">\n", "    <path d=\"M 539.088298 188.638125 \n", "L 541.069838 188.638125 \n", "L 541.069838 188.631321 \n", "L 539.088298 188.631321 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_246\">\n", "    <path d=\"M 541.069858 188.638125 \n", "L 543.051398 188.638125 \n", "L 543.051398 188.628599 \n", "L 541.069858 188.628599 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_247\">\n", "    <path d=\"M 543.051398 188.638125 \n", "L 545.032918 188.638125 \n", "L 545.032918 188.62996 \n", "L 543.051398 188.62996 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_248\">\n", "    <path d=\"M 545.032898 188.638125 \n", "L 547.014439 188.638125 \n", "L 547.014439 188.634042 \n", "L 545.032898 188.634042 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_249\">\n", "    <path d=\"M 547.014459 188.638125 \n", "L 548.995999 188.638125 \n", "L 548.995999 188.632682 \n", "L 547.014459 188.632682 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_250\">\n", "    <path d=\"M 548.995999 188.638125 \n", "L 550.977519 188.638125 \n", "L 550.977519 188.638125 \n", "L 548.995999 188.638125 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_251\">\n", "    <path d=\"M 550.977499 188.638125 \n", "L 552.959039 188.638125 \n", "L 552.959039 188.636764 \n", "L 550.977499 188.636764 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_252\">\n", "    <path d=\"M 552.959059 188.638125 \n", "L 554.9406 188.638125 \n", "L 554.9406 188.632682 \n", "L 552.959059 188.632682 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_253\">\n", "    <path d=\"M 554.94058 188.638125 \n", "L 556.92212 188.638125 \n", "L 556.92212 188.634042 \n", "L 554.94058 188.634042 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_254\">\n", "    <path d=\"M 556.92214 188.638125 \n", "L 558.90366 188.638125 \n", "L 558.90366 188.638125 \n", "L 556.92214 188.638125 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_255\">\n", "    <path d=\"M 558.90366 188.638125 \n", "L 560.8852 188.638125 \n", "L 560.8852 188.635403 \n", "L 558.90366 188.635403 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_256\">\n", "    <path d=\"M 560.88518 188.638125 \n", "L 562.866721 188.638125 \n", "L 562.866721 188.632682 \n", "L 560.88518 188.632682 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_257\">\n", "    <path d=\"M 562.866741 188.638125 \n", "L 564.848261 188.638125 \n", "L 564.848261 188.636764 \n", "L 562.866741 188.636764 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_258\">\n", "    <path d=\"M 564.848261 188.638125 \n", "L 566.829801 188.638125 \n", "L 566.829801 188.635403 \n", "L 564.848261 188.635403 \n", "z\n", "\" clip-path=\"url(#pf738aabbeb)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_259\">\n", "    <path d=\"M 34.193438 188.638125 \n", "L 34.193438 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_260\">\n", "    <path d=\"M 592.193437 188.638125 \n", "L 592.193437 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_261\">\n", "    <path d=\"M 34.193437 188.638125 \n", "L 592.193438 188.638125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_262\">\n", "    <path d=\"M 34.193437 22.318125 \n", "L 592.193438 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"text_11\">\n", "    <!-- Dequantization -->\n", "    <g style=\"fill: #262626\" transform=\"translate(267.452812 16.318125)scale(0.12 -0.12)\">\n", "     <defs>\n", "      <path id=\"DejaVuSans-44\" d=\"M 1259 4147 \n", "L 1259 519 \n", "L 2022 519 \n", "Q 2988 519 3436 956 \n", "Q 3884 1394 3884 2338 \n", "Q 3884 3275 3436 3711 \n", "Q 2988 4147 2022 4147 \n", "L 1259 4147 \n", "z\n", "M 628 4666 \n", "L 1925 4666 \n", "Q 3281 4666 3915 4102 \n", "Q 4550 3538 4550 2338 \n", "Q 4550 1131 3912 565 \n", "Q 3275 0 1925 0 \n", "L 628 0 \n", "L 628 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-65\" d=\"M 3597 1894 \n", "L 3597 1613 \n", "L 953 1613 \n", "Q 991 1019 1311 708 \n", "Q 1631 397 2203 397 \n", "Q 2534 397 2845 478 \n", "Q 3156 559 3463 722 \n", "L 3463 178 \n", "Q 3153 47 2828 -22 \n", "Q 2503 -91 2169 -91 \n", "Q 1331 -91 842 396 \n", "Q 353 884 353 1716 \n", "Q 353 2575 817 3079 \n", "Q 1281 3584 2069 3584 \n", "Q 2775 3584 3186 3129 \n", "Q 3597 2675 3597 1894 \n", "z\n", "M 3022 2063 \n", "Q 3016 2534 2758 2815 \n", "Q 2500 3097 2075 3097 \n", "Q 1594 3097 1305 2825 \n", "Q 1016 2553 972 2059 \n", "L 3022 2063 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-71\" d=\"M 947 1747 \n", "Q 947 1113 1208 752 \n", "Q 1469 391 1925 391 \n", "Q 2381 391 2643 752 \n", "Q 2906 1113 2906 1747 \n", "Q 2906 2381 2643 2742 \n", "Q 2381 3103 1925 3103 \n", "Q 1469 3103 1208 2742 \n", "Q 947 2381 947 1747 \n", "z\n", "M 2906 525 \n", "Q 2725 213 2448 61 \n", "Q 2172 -91 1784 -91 \n", "Q 1150 -91 751 415 \n", "Q 353 922 353 1747 \n", "Q 353 2572 751 3078 \n", "Q 1150 3584 1784 3584 \n", "Q 2172 3584 2448 3432 \n", "Q 2725 3281 2906 2969 \n", "L 2906 3500 \n", "L 3481 3500 \n", "L 3481 -1331 \n", "L 2906 -1331 \n", "L 2906 525 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-75\" d=\"M 544 1381 \n", "L 544 3500 \n", "L 1119 3500 \n", "L 1119 1403 \n", "Q 1119 906 1312 657 \n", "Q 1506 409 1894 409 \n", "Q 2359 409 2629 706 \n", "Q 2900 1003 2900 1516 \n", "L 2900 3500 \n", "L 3475 3500 \n", "L 3475 0 \n", "L 2900 0 \n", "L 2900 538 \n", "Q 2691 219 2414 64 \n", "Q 2138 -91 1772 -91 \n", "Q 1169 -91 856 284 \n", "Q 544 659 544 1381 \n", "z\n", "M 1991 3584 \n", "L 1991 3584 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-61\" d=\"M 2194 1759 \n", "Q 1497 1759 1228 1600 \n", "Q 959 1441 959 1056 \n", "Q 959 750 1161 570 \n", "Q 1363 391 1709 391 \n", "Q 2188 391 2477 730 \n", "Q 2766 1069 2766 1631 \n", "L 2766 1759 \n", "L 2194 1759 \n", "z\n", "M 3341 1997 \n", "L 3341 0 \n", "L 2766 0 \n", "L 2766 531 \n", "Q 2569 213 2275 61 \n", "Q 1981 -91 1556 -91 \n", "Q 1019 -91 701 211 \n", "Q 384 513 384 1019 \n", "Q 384 1609 779 1909 \n", "Q 1175 2209 1959 2209 \n", "L 2766 2209 \n", "L 2766 2266 \n", "Q 2766 2663 2505 2880 \n", "Q 2244 3097 1772 3097 \n", "Q 1472 3097 1187 3025 \n", "Q 903 2953 641 2809 \n", "L 641 3341 \n", "Q 956 3463 1253 3523 \n", "Q 1550 3584 1831 3584 \n", "Q 2591 3584 2966 3190 \n", "Q 3341 2797 3341 1997 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6e\" d=\"M 3513 2113 \n", "L 3513 0 \n", "L 2938 0 \n", "L 2938 2094 \n", "Q 2938 2591 2744 2837 \n", "Q 2550 3084 2163 3084 \n", "Q 1697 3084 1428 2787 \n", "Q 1159 2491 1159 1978 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 3500 \n", "L 1159 3500 \n", "L 1159 2956 \n", "Q 1366 3272 1645 3428 \n", "Q 1925 3584 2291 3584 \n", "Q 2894 3584 3203 3211 \n", "Q 3513 2838 3513 2113 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-74\" d=\"M 1172 4494 \n", "L 1172 3500 \n", "L 2356 3500 \n", "L 2356 3053 \n", "L 1172 3053 \n", "L 1172 1153 \n", "Q 1172 725 1289 603 \n", "Q 1406 481 1766 481 \n", "L 2356 481 \n", "L 2356 0 \n", "L 1766 0 \n", "Q 1100 0 847 248 \n", "Q 594 497 594 1153 \n", "L 594 3053 \n", "L 172 3053 \n", "L 172 3500 \n", "L 594 3500 \n", "L 594 4494 \n", "L 1172 4494 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-69\" d=\"M 603 3500 \n", "L 1178 3500 \n", "L 1178 0 \n", "L 603 0 \n", "L 603 3500 \n", "z\n", "M 603 4863 \n", "L 1178 4863 \n", "L 1178 4134 \n", "L 603 4134 \n", "L 603 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-7a\" d=\"M 353 3500 \n", "L 3084 3500 \n", "L 3084 2975 \n", "L 922 459 \n", "L 3084 459 \n", "L 3084 0 \n", "L 275 0 \n", "L 275 525 \n", "L 2438 3041 \n", "L 353 3041 \n", "L 353 3500 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6f\" d=\"M 1959 3097 \n", "Q 1497 3097 1228 2736 \n", "Q 959 2375 959 1747 \n", "Q 959 1119 1226 758 \n", "Q 1494 397 1959 397 \n", "Q 2419 397 2687 759 \n", "Q 2956 1122 2956 1747 \n", "Q 2956 2369 2687 2733 \n", "Q 2419 3097 1959 3097 \n", "z\n", "M 1959 3584 \n", "Q 2709 3584 3137 3096 \n", "Q 3566 2609 3566 1747 \n", "Q 3566 888 3137 398 \n", "Q 2709 -91 1959 -91 \n", "Q 1206 -91 779 398 \n", "Q 353 888 353 1747 \n", "Q 353 2609 779 3096 \n", "Q 1206 3584 1959 3584 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "     </defs>\n", "     <use xlink:href=\"#DejaVuSans-44\"/>\n", "     <use xlink:href=\"#DejaVuSans-65\" x=\"77.001953\"/>\n", "     <use xlink:href=\"#DejaVuSans-71\" x=\"138.525391\"/>\n", "     <use xlink:href=\"#DejaVuSans-75\" x=\"202.001953\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"265.380859\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"326.660156\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"390.039062\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"429.248047\"/>\n", "     <use xlink:href=\"#DejaVuSans-7a\" x=\"457.03125\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"509.521484\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"570.800781\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"610.009766\"/>\n", "     <use xlink:href=\"#DejaVuSans-6f\" x=\"637.792969\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"698.974609\"/>\n", "    </g>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"pf738aabbeb\">\n", "   <rect x=\"34.193438\" y=\"22.318125\" width=\"558\" height=\"166.32\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 1000x300 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["visualize_dequant_distribution(flow_dict[\"simple\"][\"model\"], sample_imgs, title=\"Dequantization\")"]}, {"cell_type": "code", "execution_count": 35, "id": "58bb60f6", "metadata": {"execution": {"iopub.execute_input": "2022-05-12T12:19:19.861772Z", "iopub.status.busy": "2022-05-12T12:19:19.861328Z", "iopub.status.idle": "2022-05-12T12:19:21.203466Z", "shell.execute_reply": "2022-05-12T12:19:21.202691Z"}, "papermill": {"duration": 1.371959, "end_time": "2022-05-12T12:19:21.206266", "exception": false, "start_time": "2022-05-12T12:19:19.834307", "status": "completed"}, "tags": []}, "outputs": [{"data": {"application/vnd.jupyter.widget-view+json": {"model_id": "d60486171a554fbfb03474a308153d0d", "version_major": 2, "version_minor": 0}, "text/plain": ["  0%|          | 0/8 [00:00<?, ?it/s]"]}, "metadata": {}, "output_type": "display_data"}, {"data": {"application/pdf": "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\n", "image/svg+xml": ["<?xml version=\"1.0\" encoding=\"utf-8\" standalone=\"no\"?>\n", "<!DOCTYPE svg PUBLIC \"-//W3C//DTD SVG 1.1//EN\"\n", "  \"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd\">\n", "<svg xmlns:xlink=\"http://www.w3.org/1999/xlink\" width=\"599.393438pt\" height=\"215.984063pt\" viewBox=\"0 0 599.393438 215.984063\" xmlns=\"http://www.w3.org/2000/svg\" version=\"1.1\">\n", " <metadata>\n", "  <rdf:RDF xmlns:dc=\"http://purl.org/dc/elements/1.1/\" xmlns:cc=\"http://creativecommons.org/ns#\" xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\">\n", "   <cc:Work>\n", "    <dc:type rdf:resource=\"http://purl.org/dc/dcmitype/StillImage\"/>\n", "    <dc:date>2022-05-12T14:19:20.625605</dc:date>\n", "    <dc:format>image/svg+xml</dc:format>\n", "    <dc:creator>\n", "     <cc:Agent>\n", "      <dc:title>Matplotlib v3.5.2, https://matplotlib.org/</dc:title>\n", "     </cc:Agent>\n", "    </dc:creator>\n", "   </cc:Work>\n", "  </rdf:RDF>\n", " </metadata>\n", " <defs>\n", "  <style type=\"text/css\">*{stroke-linejoin: round; stroke-linecap: butt}</style>\n", " </defs>\n", " <g id=\"figure_1\">\n", "  <g id=\"patch_1\">\n", "   <path d=\"M 0 215.984063 \n", "L 599.393438 215.984063 \n", "L 599.393438 0 \n", "L 0 0 \n", "z\n", "\" style=\"fill: #ffffff\"/>\n", "  </g>\n", "  <g id=\"axes_1\">\n", "   <g id=\"patch_2\">\n", "    <path d=\"M 34.193438 188.638125 \n", "L 592.193437 188.638125 \n", "L 592.193437 22.318125 \n", "L 34.193438 22.318125 \n", "z\n", "\" style=\"fill: #eaeaf2\"/>\n", "   </g>\n", "   <g id=\"matplotlib.axis_1\">\n", "    <g id=\"xtick_1\">\n", "     <g id=\"line2d_1\">\n", "      <path d=\"M 105.49848 188.638125 \n", "L 105.49848 22.318125 \n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_1\">\n", "      <!-- \u221210 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(93.890902 206.496406)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-2212\" d=\"M 678 2272 \n", "L 4684 2272 \n", "L 4684 1741 \n", "L 678 1741 \n", "L 678 2272 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-31\" d=\"M 794 531 \n", "L 1825 531 \n", "L 1825 4091 \n", "L 703 3866 \n", "L 703 4441 \n", "L 1819 4666 \n", "L 2450 4666 \n", "L 2450 531 \n", "L 3481 531 \n", "L 3481 0 \n", "L 794 0 \n", "L 794 531 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "        <path id=\"DejaVuSans-30\" d=\"M 2034 4250 \n", "Q 1547 4250 1301 3770 \n", "Q 1056 3291 1056 2328 \n", "Q 1056 1369 1301 889 \n", "Q 1547 409 2034 409 \n", "Q 2525 409 2770 889 \n", "Q 3016 1369 3016 2328 \n", "Q 3016 3291 2770 3770 \n", "Q 2525 4250 2034 4250 \n", "z\n", "M 2034 4750 \n", "Q 2819 4750 3233 4129 \n", "Q 3647 3509 3647 2328 \n", "Q 3647 1150 3233 529 \n", "Q 2819 -91 2034 -91 \n", "Q 1250 -91 836 529 \n", "Q 422 1150 422 2328 \n", "Q 422 3509 836 4129 \n", "Q 1250 4750 2034 4750 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"83.789062\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"147.412109\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_2\">\n", "     <g id=\"line2d_2\">\n", "      <path d=\"M 209.755714 188.638125 \n", "L 209.755714 22.318125 \n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_2\">\n", "      <!-- \u22125 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(201.647511 206.496406)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-35\" d=\"M 691 4666 \n", "L 3169 4666 \n", "L 3169 4134 \n", "L 1269 4134 \n", "L 1269 2991 \n", "Q 1406 3038 1543 3061 \n", "Q 1681 3084 1819 3084 \n", "Q 2600 3084 3056 2656 \n", "Q 3513 2228 3513 1497 \n", "Q 3513 744 3044 326 \n", "Q 2575 -91 1722 -91 \n", "Q 1428 -91 1123 -41 \n", "Q 819 9 494 109 \n", "L 494 744 \n", "Q 775 591 1075 516 \n", "Q 1375 441 1709 441 \n", "Q 2250 441 2565 725 \n", "Q 2881 1009 2881 1497 \n", "Q 2881 1984 2565 2268 \n", "Q 2250 2553 1709 2553 \n", "Q 1456 2553 1204 2497 \n", "Q 953 2441 691 2322 \n", "L 691 4666 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-2212\"/>\n", "       <use xlink:href=\"#DejaVuSans-35\" x=\"83.789062\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_3\">\n", "     <g id=\"line2d_3\">\n", "      <path d=\"M 314.012948 188.638125 \n", "L 314.012948 22.318125 \n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_3\">\n", "      <!-- 0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(310.513573 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_4\">\n", "     <g id=\"line2d_4\">\n", "      <path d=\"M 418.270182 188.638125 \n", "L 418.270182 22.318125 \n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_4\">\n", "      <!-- 5 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(414.770807 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-35\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"xtick_5\">\n", "     <g id=\"line2d_5\">\n", "      <path d=\"M 522.527417 188.638125 \n", "L 522.527417 22.318125 \n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_5\">\n", "      <!-- 10 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(515.528667 206.496406)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-31\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"63.623047\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"matplotlib.axis_2\">\n", "    <g id=\"ytick_1\">\n", "     <g id=\"line2d_6\">\n", "      <path d=\"M 34.193438 188.638125 \n", "L 592.193437 188.638125 \n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_6\">\n", "      <!-- 0.0 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(7.2 192.817266)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-2e\" d=\"M 684 794 \n", "L 1344 794 \n", "L 1344 0 \n", "L 684 0 \n", "L 684 794 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-30\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_2\">\n", "     <g id=\"line2d_7\">\n", "      <path d=\"M 34.193438 146.915706 \n", "L 592.193437 146.915706 \n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_7\">\n", "      <!-- 0.1 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(7.2 151.094847)scale(0.11 -0.11)\">\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-31\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_3\">\n", "     <g id=\"line2d_8\">\n", "      <path d=\"M 34.193438 105.193288 \n", "L 592.193437 105.193288 \n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_8\">\n", "      <!-- 0.2 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(7.2 109.372428)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-32\" d=\"M 1228 531 \n", "L 3431 531 \n", "L 3431 0 \n", "L 469 0 \n", "L 469 531 \n", "Q 828 903 1448 1529 \n", "Q 2069 2156 2228 2338 \n", "Q 2531 2678 2651 2914 \n", "Q 2772 3150 2772 3378 \n", "Q 2772 3750 2511 3984 \n", "Q 2250 4219 1831 4219 \n", "Q 1534 4219 1204 4116 \n", "Q 875 4013 500 3803 \n", "L 500 4441 \n", "Q 881 4594 1212 4672 \n", "Q 1544 4750 1819 4750 \n", "Q 2544 4750 2975 4387 \n", "Q 3406 4025 3406 3419 \n", "Q 3406 3131 3298 2873 \n", "Q 3191 2616 2906 2266 \n", "Q 2828 2175 2409 1742 \n", "Q 1991 1309 1228 531 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-32\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "    <g id=\"ytick_4\">\n", "     <g id=\"line2d_9\">\n", "      <path d=\"M 34.193438 63.470869 \n", "L 592.193437 63.470869 \n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: none; stroke: #ffffff; stroke-linecap: round\"/>\n", "     </g>\n", "     <g id=\"text_9\">\n", "      <!-- 0.3 -->\n", "      <g style=\"fill: #262626\" transform=\"translate(7.2 67.65001)scale(0.11 -0.11)\">\n", "       <defs>\n", "        <path id=\"DejaVuSans-33\" d=\"M 2597 2516 \n", "Q 3050 2419 3304 2112 \n", "Q 3559 1806 3559 1356 \n", "Q 3559 666 3084 287 \n", "Q 2609 -91 1734 -91 \n", "Q 1441 -91 1130 -33 \n", "Q 819 25 488 141 \n", "L 488 750 \n", "Q 750 597 1062 519 \n", "Q 1375 441 1716 441 \n", "Q 2309 441 2620 675 \n", "Q 2931 909 2931 1356 \n", "Q 2931 1769 2642 2001 \n", "Q 2353 2234 1838 2234 \n", "L 1294 2234 \n", "L 1294 2753 \n", "L 1863 2753 \n", "Q 2328 2753 2575 2939 \n", "Q 2822 3125 2822 3475 \n", "Q 2822 3834 2567 4026 \n", "Q 2313 4219 1838 4219 \n", "Q 1578 4219 1281 4162 \n", "Q 984 4106 628 3988 \n", "L 628 4550 \n", "Q 988 4650 1302 4700 \n", "Q 1616 4750 1894 4750 \n", "Q 2613 4750 3031 4423 \n", "Q 3450 4097 3450 3541 \n", "Q 3450 3153 3228 2886 \n", "Q 3006 2619 2597 2516 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "       </defs>\n", "       <use xlink:href=\"#DejaVuSans-30\"/>\n", "       <use xlink:href=\"#DejaVuSans-2e\" x=\"63.623047\"/>\n", "       <use xlink:href=\"#DejaVuSans-33\" x=\"95.410156\"/>\n", "      </g>\n", "     </g>\n", "    </g>\n", "   </g>\n", "   <g id=\"patch_3\">\n", "    <path d=\"M 59.557074 188.638125 \n", "L 61.538603 188.638125 \n", "L 61.538603 188.474062 \n", "L 59.557074 188.474062 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_4\">\n", "    <path d=\"M 61.538623 188.638125 \n", "L 63.520152 188.638125 \n", "L 63.520152 188.312734 \n", "L 61.538623 188.312734 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_5\">\n", "    <path d=\"M 63.520152 188.638125 \n", "L 65.501702 188.638125 \n", "L 65.501702 188.241644 \n", "L 63.520152 188.241644 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_6\">\n", "    <path d=\"M 65.501682 188.638125 \n", "L 67.483211 188.638125 \n", "L 67.483211 188.099452 \n", "L 65.501682 188.099452 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_7\">\n", "    <path d=\"M 67.483231 188.638125 \n", "L 69.46476 188.638125 \n", "L 69.46476 187.888904 \n", "L 67.483231 187.888904 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_8\">\n", "    <path d=\"M 69.464741 188.638125 \n", "L 71.44627 188.638125 \n", "L 71.44627 187.634607 \n", "L 69.464741 187.634607 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_9\">\n", "    <path d=\"M 71.44629 188.638125 \n", "L 73.427839 188.638125 \n", "L 73.427839 187.235401 \n", "L 71.44629 187.235401 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_10\">\n", "    <path d=\"M 73.427839 188.638125 \n", "L 75.409368 188.638125 \n", "L 75.409368 186.874449 \n", "L 73.427839 186.874449 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_11\">\n", "    <path d=\"M 75.409348 188.638125 \n", "L 77.390878 188.638125 \n", "L 77.390878 186.267416 \n", "L 75.409348 186.267416 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_12\">\n", "    <path d=\"M 77.390898 188.638125 \n", "L 79.372427 188.638125 \n", "L 79.372427 185.611164 \n", "L 77.390898 185.611164 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_13\">\n", "    <path d=\"M 79.372427 188.638125 \n", "L 81.353976 188.638125 \n", "L 81.353976 184.883857 \n", "L 79.372427 184.883857 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_14\">\n", "    <path d=\"M 81.353956 188.638125 \n", "L 83.335486 188.638125 \n", "L 83.335486 183.970535 \n", "L 81.353956 183.970535 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_15\">\n", "    <path d=\"M 83.335506 188.638125 \n", "L 85.317035 188.638125 \n", "L 85.317035 182.524047 \n", "L 83.335506 182.524047 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_16\">\n", "    <path d=\"M 85.317015 188.638125 \n", "L 87.298544 188.638125 \n", "L 87.298544 181.293576 \n", "L 85.317015 181.293576 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_17\">\n", "    <path d=\"M 87.298564 188.638125 \n", "L 89.280114 188.638125 \n", "L 89.280114 179.478038 \n", "L 87.298564 179.478038 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_18\">\n", "    <path d=\"M 89.280114 188.638125 \n", "L 91.261643 188.638125 \n", "L 91.261643 177.112706 \n", "L 89.280114 177.112706 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_19\">\n", "    <path d=\"M 91.261623 188.638125 \n", "L 93.243152 188.638125 \n", "L 93.243152 174.640825 \n", "L 91.261623 174.640825 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_20\">\n", "    <path d=\"M 93.243172 188.638125 \n", "L 95.224702 188.638125 \n", "L 95.224702 171.613864 \n", "L 93.243172 171.613864 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_21\">\n", "    <path d=\"M 95.224682 188.638125 \n", "L 97.206211 188.638125 \n", "L 97.206211 167.679089 \n", "L 95.224682 167.679089 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_22\">\n", "    <path d=\"M 97.206231 188.638125 \n", "L 99.18778 188.638125 \n", "L 99.18778 162.812149 \n", "L 97.206231 162.812149 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_23\">\n", "    <path d=\"M 99.18778 188.638125 \n", "L 101.16931 188.638125 \n", "L 101.16931 157.78883 \n", "L 99.18778 157.78883 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_24\">\n", "    <path d=\"M 101.16929 188.638125 \n", "L 103.150819 188.638125 \n", "L 103.150819 152.607177 \n", "L 101.16929 152.607177 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_25\">\n", "    <path d=\"M 103.150839 188.638125 \n", "L 105.132368 188.638125 \n", "L 105.132368 145.475909 \n", "L 103.150839 145.475909 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_26\">\n", "    <path d=\"M 105.132368 188.638125 \n", "L 107.113918 188.638125 \n", "L 107.113918 137.943196 \n", "L 105.132368 137.943196 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_27\">\n", "    <path d=\"M 107.113898 188.638125 \n", "L 109.095427 188.638125 \n", "L 109.095427 129.545401 \n", "L 107.113898 129.545401 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_28\">\n", "    <path d=\"M 109.095447 188.638125 \n", "L 111.076976 188.638125 \n", "L 111.076976 120.65866 \n", "L 109.095447 120.65866 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_29\">\n", "    <path d=\"M 111.076956 188.638125 \n", "L 113.058486 188.638125 \n", "L 113.058486 111.74731 \n", "L 111.076956 111.74731 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_30\">\n", "    <path d=\"M 113.058506 188.638125 \n", "L 115.040055 188.638125 \n", "L 115.040055 101.877063 \n", "L 113.058506 101.877063 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_31\">\n", "    <path d=\"M 115.040055 188.638125 \n", "L 117.021584 188.638125 \n", "L 117.021584 91.318744 \n", "L 115.040055 91.318744 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_32\">\n", "    <path d=\"M 117.021564 188.638125 \n", "L 119.003094 188.638125 \n", "L 119.003094 80.824187 \n", "L 117.021564 80.824187 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_33\">\n", "    <path d=\"M 119.003114 188.638125 \n", "L 120.984643 188.638125 \n", "L 120.984643 70.991351 \n", "L 119.003114 70.991351 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_34\">\n", "    <path d=\"M 120.984623 188.638125 \n", "L 122.966152 188.638125 \n", "L 122.966152 60.496793 \n", "L 120.984623 60.496793 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_35\">\n", "    <path d=\"M 122.966172 188.638125 \n", "L 124.947721 188.638125 \n", "L 124.947721 52.532905 \n", "L 122.966172 52.532905 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_36\">\n", "    <path d=\"M 124.947721 188.638125 \n", "L 126.929251 188.638125 \n", "L 126.929251 44.549879 \n", "L 124.947721 44.549879 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_37\">\n", "    <path d=\"M 126.929231 188.638125 \n", "L 128.91076 188.638125 \n", "L 128.91076 38.036582 \n", "L 126.929231 38.036582 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_38\">\n", "    <path d=\"M 128.91078 188.638125 \n", "L 130.89231 188.638125 \n", "L 130.89231 34.257666 \n", "L 128.91078 34.257666 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_39\">\n", "    <path d=\"M 130.89231 188.638125 \n", "L 132.873859 188.638125 \n", "L 132.873859 31.172129 \n", "L 130.89231 31.172129 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_40\">\n", "    <path d=\"M 132.873839 188.638125 \n", "L 134.855368 188.638125 \n", "L 134.855368 30.238125 \n", "L 132.873839 30.238125 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_41\">\n", "    <path d=\"M 134.855388 188.638125 \n", "L 136.836917 188.638125 \n", "L 136.836917 32.797506 \n", "L 134.855388 32.797506 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_42\">\n", "    <path d=\"M 136.836898 188.638125 \n", "L 138.818427 188.638125 \n", "L 138.818427 37.112361 \n", "L 136.836898 37.112361 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_43\">\n", "    <path d=\"M 138.818447 188.638125 \n", "L 140.799996 188.638125 \n", "L 140.799996 43.55602 \n", "L 138.818447 43.55602 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_44\">\n", "    <path d=\"M 140.799996 188.638125 \n", "L 142.781525 188.638125 \n", "L 142.781525 50.953801 \n", "L 140.799996 50.953801 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_45\">\n", "    <path d=\"M 142.781506 188.638125 \n", "L 144.763035 188.638125 \n", "L 144.763035 59.211634 \n", "L 142.781506 59.211634 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_46\">\n", "    <path d=\"M 144.763055 188.638125 \n", "L 146.744584 188.638125 \n", "L 146.744584 68.519469 \n", "L 144.763055 68.519469 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_47\">\n", "    <path d=\"M 146.744584 188.638125 \n", "L 148.726133 188.638125 \n", "L 148.726133 80.411839 \n", "L 146.744584 80.411839 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_48\">\n", "    <path d=\"M 148.726113 188.638125 \n", "L 150.707663 188.638125 \n", "L 150.707663 90.381344 \n", "L 148.726113 90.381344 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_49\">\n", "    <path d=\"M 150.707663 188.638125 \n", "L 152.689192 188.638125 \n", "L 152.689192 102.070333 \n", "L 150.707663 102.070333 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_50\">\n", "    <path d=\"M 152.689192 188.638125 \n", "L 154.670721 188.638125 \n", "L 154.670721 111.818404 \n", "L 152.689192 111.818404 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_51\">\n", "    <path d=\"M 154.670721 188.638125 \n", "L 156.652271 188.638125 \n", "L 156.652271 122.627747 \n", "L 154.670721 122.627747 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_52\">\n", "    <path d=\"M 156.652261 188.638125 \n", "L 158.63379 188.638125 \n", "L 158.63379 131.880563 \n", "L 156.652261 131.880563 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_53\">\n", "    <path d=\"M 158.63378 188.638125 \n", "L 160.615329 188.638125 \n", "L 160.615329 140.187858 \n", "L 158.63378 140.187858 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_54\">\n", "    <path d=\"M 160.615329 188.638125 \n", "L 162.596859 188.638125 \n", "L 162.596859 147.414586 \n", "L 160.615329 147.414586 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_55\">\n", "    <path d=\"M 162.596859 188.638125 \n", "L 164.578408 188.638125 \n", "L 164.578408 154.461259 \n", "L 162.596859 154.461259 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_56\">\n", "    <path d=\"M 164.578398 188.638125 \n", "L 166.559927 188.638125 \n", "L 166.559927 159.566178 \n", "L 164.578398 159.566178 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_57\">\n", "    <path d=\"M 166.559917 188.638125 \n", "L 168.541467 188.638125 \n", "L 168.541467 164.77803 \n", "L 166.559917 164.77803 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_58\">\n", "    <path d=\"M 168.541467 188.638125 \n", "L 170.522996 188.638125 \n", "L 170.522996 168.720888 \n", "L 168.541467 168.720888 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_59\">\n", "    <path d=\"M 170.522996 188.638125 \n", "L 172.504545 188.638125 \n", "L 172.504545 172.573713 \n", "L 170.522996 172.573713 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_60\">\n", "    <path d=\"M 172.504535 188.638125 \n", "L 174.486065 188.638125 \n", "L 174.486065 175.075592 \n", "L 172.504535 175.075592 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_61\">\n", "    <path d=\"M 174.486055 188.638125 \n", "L 176.467604 188.638125 \n", "L 176.467604 177.730731 \n", "L 174.486055 177.730731 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_62\">\n", "    <path d=\"M 176.467604 188.638125 \n", "L 178.449133 188.638125 \n", "L 178.449133 179.467009 \n", "L 176.467604 179.467009 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_63\">\n", "    <path d=\"M 178.449133 188.638125 \n", "L 180.430683 188.638125 \n", "L 180.430683 181.036582 \n", "L 178.449133 181.036582 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_64\">\n", "    <path d=\"M 180.430673 188.638125 \n", "L 182.412202 188.638125 \n", "L 182.412202 182.242406 \n", "L 180.430673 182.242406 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_65\">\n", "    <path d=\"M 182.412192 188.638125 \n", "L 184.393741 188.638125 \n", "L 184.393741 183.308842 \n", "L 182.412192 183.308842 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_66\">\n", "    <path d=\"M 184.393741 188.638125 \n", "L 186.375271 188.638125 \n", "L 186.375271 184.219364 \n", "L 184.393741 184.219364 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_67\">\n", "    <path d=\"M 186.375271 188.638125 \n", "L 188.3568 188.638125 \n", "L 188.3568 184.872881 \n", "L 186.375271 184.872881 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_68\">\n", "    <path d=\"M 188.3568 188.638125 \n", "L 190.338349 188.638125 \n", "L 190.338349 185.378758 \n", "L 188.3568 185.378758 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_69\">\n", "    <path d=\"M 190.338339 188.638125 \n", "L 192.319869 188.638125 \n", "L 192.319869 185.865462 \n", "L 190.338339 185.865462 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_70\">\n", "    <path d=\"M 192.319859 188.638125 \n", "L 194.301408 188.638125 \n", "L 194.301408 186.316647 \n", "L 192.319859 186.316647 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_71\">\n", "    <path d=\"M 194.301408 188.638125 \n", "L 196.282937 188.638125 \n", "L 196.282937 186.639292 \n", "L 194.301408 186.639292 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_72\">\n", "    <path d=\"M 196.282937 188.638125 \n", "L 198.264486 188.638125 \n", "L 198.264486 187.08227 \n", "L 196.282937 187.08227 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_73\">\n", "    <path d=\"M 198.264477 188.638125 \n", "L 200.246006 188.638125 \n", "L 200.246006 188.230702 \n", "L 198.264477 188.230702 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_74\">\n", "    <path d=\"M 200.245996 188.638125 \n", "L 202.227545 188.638125 \n", "L 202.227545 188.37836 \n", "L 200.245996 188.37836 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_75\">\n", "    <path d=\"M 202.227545 188.638125 \n", "L 204.209075 188.638125 \n", "L 204.209075 188.438515 \n", "L 202.227545 188.438515 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_76\">\n", "    <path d=\"M 204.209075 188.638125 \n", "L 206.190624 188.638125 \n", "L 206.190624 188.405704 \n", "L 204.209075 188.405704 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_77\">\n", "    <path d=\"M 206.190614 188.638125 \n", "L 208.172143 188.638125 \n", "L 208.172143 188.424843 \n", "L 206.190614 188.424843 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_78\">\n", "    <path d=\"M 208.172133 188.638125 \n", "L 210.153682 188.638125 \n", "L 210.153682 188.498672 \n", "L 208.172133 188.498672 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_79\">\n", "    <path d=\"M 210.153682 188.638125 \n", "L 212.135212 188.638125 \n", "L 212.135212 188.463125 \n", "L 210.153682 188.463125 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_80\">\n", "    <path d=\"M 212.135212 188.638125 \n", "L 214.116761 188.638125 \n", "L 214.116761 188.143204 \n", "L 212.135212 188.143204 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_81\">\n", "    <path d=\"M 214.116751 188.638125 \n", "L 216.09828 188.638125 \n", "L 216.09828 187.905311 \n", "L 214.116751 187.905311 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_82\">\n", "    <path d=\"M 216.098271 188.638125 \n", "L 218.07982 188.638125 \n", "L 218.07982 187.910783 \n", "L 216.098271 187.910783 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_83\">\n", "    <path d=\"M 218.07982 188.638125 \n", "L 220.061349 188.638125 \n", "L 220.061349 187.782264 \n", "L 218.07982 187.782264 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_84\">\n", "    <path d=\"M 220.061349 188.638125 \n", "L 222.042878 188.638125 \n", "L 222.042878 187.779529 \n", "L 220.061349 187.779529 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_85\">\n", "    <path d=\"M 222.042878 188.638125 \n", "L 224.024428 188.638125 \n", "L 224.024428 187.864299 \n", "L 222.042878 187.864299 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_86\">\n", "    <path d=\"M 224.024418 188.638125 \n", "L 226.005947 188.638125 \n", "L 226.005947 187.689295 \n", "L 224.024418 187.689295 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_87\">\n", "    <path d=\"M 226.005937 188.638125 \n", "L 227.987486 188.638125 \n", "L 227.987486 187.517034 \n", "L 226.005937 187.517034 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_88\">\n", "    <path d=\"M 227.987486 188.638125 \n", "L 229.969016 188.638125 \n", "L 229.969016 187.183434 \n", "L 227.987486 187.183434 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_89\">\n", "    <path d=\"M 229.969016 188.638125 \n", "L 231.950565 188.638125 \n", "L 231.950565 187.208051 \n", "L 229.969016 187.208051 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_90\">\n", "    <path d=\"M 231.950555 188.638125 \n", "L 233.932084 188.638125 \n", "L 233.932084 187.221715 \n", "L 231.950555 187.221715 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_91\">\n", "    <path d=\"M 233.932084 188.638125 \n", "L 235.913614 188.638125 \n", "L 235.913614 187.112344 \n", "L 233.932084 187.112344 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_92\">\n", "    <path d=\"M 235.913624 188.638125 \n", "L 237.895153 188.638125 \n", "L 237.895153 186.406875 \n", "L 235.913624 186.406875 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_93\">\n", "    <path d=\"M 237.895153 188.638125 \n", "L 239.876682 188.638125 \n", "L 239.876682 186.220938 \n", "L 237.895153 186.220938 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_94\">\n", "    <path d=\"M 239.876692 188.638125 \n", "L 241.858222 188.638125 \n", "L 241.858222 186.401406 \n", "L 239.876692 186.401406 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_95\">\n", "    <path d=\"M 241.858222 188.638125 \n", "L 243.839751 188.638125 \n", "L 243.839751 186.275625 \n", "L 241.858222 186.275625 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_96\">\n", "    <path d=\"M 243.839761 188.638125 \n", "L 245.82129 188.638125 \n", "L 245.82129 185.734219 \n", "L 243.839761 185.734219 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_97\">\n", "    <path d=\"M 245.82129 188.638125 \n", "L 247.80282 188.638125 \n", "L 247.80282 185.042422 \n", "L 245.82129 185.042422 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_98\">\n", "    <path d=\"M 247.80283 188.638125 \n", "L 249.784359 188.638125 \n", "L 249.784359 185.138125 \n", "L 247.80283 185.138125 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_99\">\n", "    <path d=\"M 249.784359 188.638125 \n", "L 251.765888 188.638125 \n", "L 251.765888 185.00961 \n", "L 249.784359 185.00961 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_100\">\n", "    <path d=\"M 251.765898 188.638125 \n", "L 253.747428 188.638125 \n", "L 253.747428 185.088906 \n", "L 251.765898 185.088906 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_101\">\n", "    <path d=\"M 253.747428 188.638125 \n", "L 255.728957 188.638125 \n", "L 255.728957 184.684219 \n", "L 253.747428 184.684219 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_102\">\n", "    <path d=\"M 255.728967 188.638125 \n", "L 257.710496 188.638125 \n", "L 257.710496 184.943985 \n", "L 255.728967 184.943985 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_103\">\n", "    <path d=\"M 257.710496 188.638125 \n", "L 259.692026 188.638125 \n", "L 259.692026 184.924844 \n", "L 257.710496 184.924844 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_104\">\n", "    <path d=\"M 259.692036 188.638125 \n", "L 261.673565 188.638125 \n", "L 261.673565 184.205703 \n", "L 259.692036 184.205703 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_105\">\n", "    <path d=\"M 261.673565 188.638125 \n", "L 263.655094 188.638125 \n", "L 263.655094 183.549453 \n", "L 261.673565 183.549453 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_106\">\n", "    <path d=\"M 263.655104 188.638125 \n", "L 265.636634 188.638125 \n", "L 265.636634 182.693594 \n", "L 263.655104 182.693594 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_107\">\n", "    <path d=\"M 265.636634 188.638125 \n", "L 267.618163 188.638125 \n", "L 267.618163 184.123661 \n", "L 265.636634 184.123661 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_108\">\n", "    <path d=\"M 267.618163 188.638125 \n", "L 269.599692 188.638125 \n", "L 269.599692 183.623281 \n", "L 267.618163 183.623281 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_109\">\n", "    <path d=\"M 269.599702 188.638125 \n", "L 271.581232 188.638125 \n", "L 271.581232 182.480313 \n", "L 269.599702 182.480313 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_110\">\n", "    <path d=\"M 271.581232 188.638125 \n", "L 273.562761 188.638125 \n", "L 273.562761 181.413907 \n", "L 271.581232 181.413907 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_111\">\n", "    <path d=\"M 273.562768 188.638125 \n", "L 275.544298 188.638125 \n", "L 275.544298 183.344375 \n", "L 273.562768 183.344375 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_112\">\n", "    <path d=\"M 275.544303 188.638125 \n", "L 277.525832 188.638125 \n", "L 277.525832 182.070157 \n", "L 275.544303 182.070157 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_113\">\n", "    <path d=\"M 277.525837 188.638125 \n", "L 279.507366 188.638125 \n", "L 279.507366 182.007266 \n", "L 277.525837 182.007266 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_114\">\n", "    <path d=\"M 279.507371 188.638125 \n", "L 281.488901 188.638125 \n", "L 281.488901 181.156875 \n", "L 279.507371 181.156875 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_115\">\n", "    <path d=\"M 281.488906 188.638125 \n", "L 283.470435 188.638125 \n", "L 283.470435 182.20961 \n", "L 281.488906 182.20961 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_116\">\n", "    <path d=\"M 283.47044 188.638125 \n", "L 285.451969 188.638125 \n", "L 285.451969 181.135 \n", "L 283.47044 181.135 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_117\">\n", "    <path d=\"M 285.451974 188.638125 \n", "L 287.433504 188.638125 \n", "L 287.433504 182.362735 \n", "L 285.451974 182.362735 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_118\">\n", "    <path d=\"M 287.433509 188.638125 \n", "L 289.415038 188.638125 \n", "L 289.415038 180.191641 \n", "L 287.433509 180.191641 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_119\">\n", "    <path d=\"M 289.415041 188.638125 \n", "L 291.39657 188.638125 \n", "L 291.39657 178.723269 \n", "L 289.415041 178.723269 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_120\">\n", "    <path d=\"M 291.396575 188.638125 \n", "L 293.378104 188.638125 \n", "L 293.378104 180.292813 \n", "L 291.396575 180.292813 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_121\">\n", "    <path d=\"M 293.378109 188.638125 \n", "L 295.359639 188.638125 \n", "L 295.359639 180.071328 \n", "L 293.378109 180.071328 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_122\">\n", "    <path d=\"M 295.359643 188.638125 \n", "L 297.341173 188.638125 \n", "L 297.341173 179.606485 \n", "L 295.359643 179.606485 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_123\">\n", "    <path d=\"M 297.341178 188.638125 \n", "L 299.322707 188.638125 \n", "L 299.322707 179.652969 \n", "L 297.341178 179.652969 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_124\">\n", "    <path d=\"M 299.322712 188.638125 \n", "L 301.304241 188.638125 \n", "L 301.304241 177.875625 \n", "L 299.322712 177.875625 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_125\">\n", "    <path d=\"M 301.304246 188.638125 \n", "L 303.285776 188.638125 \n", "L 303.285776 180.123276 \n", "L 301.304246 180.123276 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_126\">\n", "    <path d=\"M 303.28578 188.638125 \n", "L 305.267309 188.638125 \n", "L 305.267309 179.669375 \n", "L 303.28578 179.669375 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_127\">\n", "    <path d=\"M 305.267314 188.638125 \n", "L 307.248843 188.638125 \n", "L 307.248843 179.212735 \n", "L 305.267314 179.212735 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_128\">\n", "    <path d=\"M 307.248848 188.638125 \n", "L 309.230378 188.638125 \n", "L 309.230378 179.37406 \n", "L 307.248848 179.37406 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_129\">\n", "    <path d=\"M 309.230382 188.638125 \n", "L 311.211911 188.638125 \n", "L 311.211911 178.392421 \n", "L 309.230382 178.392421 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_130\">\n", "    <path d=\"M 311.211916 188.638125 \n", "L 313.193445 188.638125 \n", "L 313.193445 178.602968 \n", "L 311.211916 178.602968 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_131\">\n", "    <path d=\"M 313.19345 188.638125 \n", "L 315.174979 188.638125 \n", "L 315.174979 173.44867 \n", "L 313.19345 173.44867 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_132\">\n", "    <path d=\"M 315.174984 188.638125 \n", "L 317.156513 188.638125 \n", "L 317.156513 179.05414 \n", "L 315.174984 179.05414 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_133\">\n", "    <path d=\"M 317.156518 188.638125 \n", "L 319.138047 188.638125 \n", "L 319.138047 177.656874 \n", "L 317.156518 177.656874 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_134\">\n", "    <path d=\"M 319.138052 188.638125 \n", "L 321.119581 188.638125 \n", "L 321.119581 178.884608 \n", "L 319.138052 178.884608 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_135\">\n", "    <path d=\"M 321.119586 188.638125 \n", "L 323.101115 188.638125 \n", "L 323.101115 178.843594 \n", "L 321.119586 178.843594 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_136\">\n", "    <path d=\"M 323.10112 188.638125 \n", "L 325.082648 188.638125 \n", "L 325.082648 180.314683 \n", "L 323.10112 180.314683 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_137\">\n", "    <path d=\"M 325.082653 188.638125 \n", "L 327.064183 188.638125 \n", "L 327.064183 180.090469 \n", "L 325.082653 180.090469 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_138\">\n", "    <path d=\"M 327.064188 188.638125 \n", "L 329.045717 188.638125 \n", "L 329.045717 176.87211 \n", "L 327.064188 176.87211 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_139\">\n", "    <path d=\"M 329.045722 188.638125 \n", "L 331.027251 188.638125 \n", "L 331.027251 179.776016 \n", "L 329.045722 179.776016 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_140\">\n", "    <path d=\"M 331.027256 188.638125 \n", "L 333.008786 188.638125 \n", "L 333.008786 178.993985 \n", "L 331.027256 178.993985 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_141\">\n", "    <path d=\"M 333.008791 188.638125 \n", "L 334.99032 188.638125 \n", "L 334.99032 181.722891 \n", "L 333.008791 181.722891 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_142\">\n", "    <path d=\"M 334.990325 188.638125 \n", "L 336.971854 188.638125 \n", "L 336.971854 176.907642 \n", "L 334.990325 176.907642 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_143\">\n", "    <path d=\"M 336.971857 188.638125 \n", "L 338.953386 188.638125 \n", "L 338.953386 180.156094 \n", "L 336.971857 180.156094 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_144\">\n", "    <path d=\"M 338.953391 188.638125 \n", "L 340.934921 188.638125 \n", "L 340.934921 179.97836 \n", "L 338.953391 179.97836 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_145\">\n", "    <path d=\"M 340.934925 188.638125 \n", "L 342.916455 188.638125 \n", "L 342.916455 181.763907 \n", "L 340.934925 181.763907 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_146\">\n", "    <path d=\"M 342.91646 188.638125 \n", "L 344.897989 188.638125 \n", "L 344.897989 181.996328 \n", "L 342.91646 181.996328 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_147\">\n", "    <path d=\"M 344.897994 188.638125 \n", "L 346.879523 188.638125 \n", "L 346.879523 181.156875 \n", "L 344.897994 181.156875 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_148\">\n", "    <path d=\"M 346.879528 188.638125 \n", "L 348.861058 188.638125 \n", "L 348.861058 180.918985 \n", "L 346.879528 180.918985 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_149\">\n", "    <path d=\"M 348.861063 188.638125 \n", "L 350.842592 188.638125 \n", "L 350.842592 181.326407 \n", "L 348.861063 181.326407 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_150\">\n", "    <path d=\"M 350.842597 188.638125 \n", "L 352.824126 188.638125 \n", "L 352.824126 183.311563 \n", "L 350.842597 183.311563 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_151\">\n", "    <path d=\"M 352.824129 188.638125 \n", "L 354.805658 188.638125 \n", "L 354.805658 181.936164 \n", "L 352.824129 181.936164 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_152\">\n", "    <path d=\"M 354.805663 188.638125 \n", "L 356.787193 188.638125 \n", "L 356.787193 181.449462 \n", "L 354.805663 181.449462 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_153\">\n", "    <path d=\"M 356.787203 188.638125 \n", "L 358.768732 188.638125 \n", "L 358.768732 183.076406 \n", "L 356.787203 183.076406 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_154\">\n", "    <path d=\"M 358.768732 188.638125 \n", "L 360.750261 188.638125 \n", "L 360.750261 183.620535 \n", "L 358.768732 183.620535 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_155\">\n", "    <path d=\"M 360.750261 188.638125 \n", "L 362.731791 188.638125 \n", "L 362.731791 182.75375 \n", "L 360.750261 182.75375 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_156\">\n", "    <path d=\"M 362.731801 188.638125 \n", "L 364.71333 188.638125 \n", "L 364.71333 183.213125 \n", "L 362.731801 183.213125 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_157\">\n", "    <path d=\"M 364.71333 188.638125 \n", "L 366.694859 188.638125 \n", "L 366.694859 183.4975 \n", "L 364.71333 183.4975 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_158\">\n", "    <path d=\"M 366.694869 188.638125 \n", "L 368.676399 188.638125 \n", "L 368.676399 184.345156 \n", "L 366.694869 184.345156 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_159\">\n", "    <path d=\"M 368.676399 188.638125 \n", "L 370.657928 188.638125 \n", "L 370.657928 184.700625 \n", "L 368.676399 184.700625 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_160\">\n", "    <path d=\"M 370.657938 188.638125 \n", "L 372.639467 188.638125 \n", "L 372.639467 184.670547 \n", "L 370.657938 184.670547 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_161\">\n", "    <path d=\"M 372.639467 188.638125 \n", "L 374.620997 188.638125 \n", "L 374.620997 184.46 \n", "L 372.639467 184.46 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_162\">\n", "    <path d=\"M 374.621007 188.638125 \n", "L 376.602536 188.638125 \n", "L 376.602536 185.099844 \n", "L 374.621007 185.099844 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_163\">\n", "    <path d=\"M 376.602536 188.638125 \n", "L 378.584065 188.638125 \n", "L 378.584065 185.389688 \n", "L 376.602536 185.389688 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_164\">\n", "    <path d=\"M 378.584075 188.638125 \n", "L 380.565605 188.638125 \n", "L 380.565605 185.165469 \n", "L 378.584075 185.165469 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_165\">\n", "    <path d=\"M 380.565605 188.638125 \n", "L 382.547134 188.638125 \n", "L 382.547134 186.215469 \n", "L 380.565605 186.215469 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_166\">\n", "    <path d=\"M 382.547144 188.638125 \n", "L 384.528673 188.638125 \n", "L 384.528673 185.881875 \n", "L 382.547144 185.881875 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_167\">\n", "    <path d=\"M 384.528673 188.638125 \n", "L 386.510203 188.638125 \n", "L 386.510203 186.155313 \n", "L 384.528673 186.155313 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_168\">\n", "    <path d=\"M 386.510212 188.638125 \n", "L 388.491742 188.638125 \n", "L 388.491742 186.385 \n", "L 386.510212 186.385 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_169\">\n", "    <path d=\"M 388.491742 188.638125 \n", "L 390.473271 188.638125 \n", "L 390.473271 186.464297 \n", "L 388.491742 186.464297 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_170\">\n", "    <path d=\"M 390.473281 188.638125 \n", "L 392.45481 188.638125 \n", "L 392.45481 186.631094 \n", "L 390.473281 186.631094 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_171\">\n", "    <path d=\"M 392.45481 188.638125 \n", "L 394.43634 188.638125 \n", "L 394.43634 186.652964 \n", "L 392.45481 186.652964 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_172\">\n", "    <path d=\"M 394.43634 188.638125 \n", "L 396.417869 188.638125 \n", "L 396.417869 186.335781 \n", "L 394.43634 186.335781 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_173\">\n", "    <path d=\"M 396.417879 188.638125 \n", "L 398.399408 188.638125 \n", "L 398.399408 183.770938 \n", "L 396.417879 183.770938 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_174\">\n", "    <path d=\"M 398.399408 188.638125 \n", "L 400.380958 188.638125 \n", "L 400.380958 184.604932 \n", "L 398.399408 184.604932 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_175\">\n", "    <path d=\"M 400.380948 188.638125 \n", "L 402.362477 188.638125 \n", "L 402.362477 166.391195 \n", "L 400.380948 166.391195 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_176\">\n", "    <path d=\"M 402.362467 188.638125 \n", "L 404.344016 188.638125 \n", "L 404.344016 160.53429 \n", "L 402.362467 160.53429 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_177\">\n", "    <path d=\"M 404.344016 188.638125 \n", "L 406.325546 188.638125 \n", "L 406.325546 168.27245 \n", "L 404.344016 168.27245 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_178\">\n", "    <path d=\"M 406.325546 188.638125 \n", "L 408.307075 188.638125 \n", "L 408.307075 164.693144 \n", "L 406.325546 164.693144 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_179\">\n", "    <path d=\"M 408.307075 188.638125 \n", "L 410.288624 188.638125 \n", "L 410.288624 155.470241 \n", "L 408.307075 155.470241 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_180\">\n", "    <path d=\"M 410.288614 188.638125 \n", "L 412.270144 188.638125 \n", "L 412.270144 156.536483 \n", "L 410.288614 156.536483 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_181\">\n", "    <path d=\"M 412.270134 188.638125 \n", "L 414.251683 188.638125 \n", "L 414.251683 165.158107 \n", "L 412.270134 165.158107 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_182\">\n", "    <path d=\"M 414.251683 188.638125 \n", "L 416.233212 188.638125 \n", "L 416.233212 177.088097 \n", "L 414.251683 177.088097 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_183\">\n", "    <path d=\"M 416.233212 188.638125 \n", "L 418.214762 188.638125 \n", "L 418.214762 174.394802 \n", "L 416.233212 174.394802 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_184\">\n", "    <path d=\"M 418.214752 188.638125 \n", "L 420.196281 188.638125 \n", "L 420.196281 173.568947 \n", "L 418.214752 173.568947 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_185\">\n", "    <path d=\"M 420.196271 188.638125 \n", "L 422.17782 188.638125 \n", "L 422.17782 174.074881 \n", "L 420.196271 174.074881 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_186\">\n", "    <path d=\"M 422.17782 188.638125 \n", "L 424.15935 188.638125 \n", "L 424.15935 175.778328 \n", "L 422.17782 175.778328 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_187\">\n", "    <path d=\"M 424.15935 188.638125 \n", "L 426.140899 188.638125 \n", "L 426.140899 177.678778 \n", "L 424.15935 177.678778 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_188\">\n", "    <path d=\"M 426.140889 188.638125 \n", "L 428.122418 188.638125 \n", "L 428.122418 180.426777 \n", "L 426.140889 180.426777 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_189\">\n", "    <path d=\"M 428.122408 188.638125 \n", "L 430.103958 188.638125 \n", "L 430.103958 184.306886 \n", "L 428.122408 184.306886 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_190\">\n", "    <path d=\"M 430.103958 188.638125 \n", "L 432.085487 188.638125 \n", "L 432.085487 186.888121 \n", "L 430.103958 186.888121 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_191\">\n", "    <path d=\"M 432.085487 188.638125 \n", "L 434.067036 188.638125 \n", "L 434.067036 186.499849 \n", "L 432.085487 186.499849 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_192\">\n", "    <path d=\"M 434.067026 188.638125 \n", "L 436.048556 188.638125 \n", "L 436.048556 186.614683 \n", "L 434.067026 186.614683 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_193\">\n", "    <path d=\"M 436.048546 188.638125 \n", "L 438.030095 188.638125 \n", "L 438.030095 186.56274 \n", "L 436.048546 186.56274 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_194\">\n", "    <path d=\"M 438.030095 188.638125 \n", "L 440.011624 188.638125 \n", "L 440.011624 186.661167 \n", "L 438.030095 186.661167 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_195\">\n", "    <path d=\"M 440.011624 188.638125 \n", "L 441.993154 188.638125 \n", "L 441.993154 186.69398 \n", "L 440.011624 186.69398 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_196\">\n", "    <path d=\"M 441.993154 188.638125 \n", "L 443.974703 188.638125 \n", "L 443.974703 186.762349 \n", "L 441.993154 186.762349 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_197\">\n", "    <path d=\"M 443.974693 188.638125 \n", "L 445.956222 188.638125 \n", "L 445.956222 186.986558 \n", "L 443.974693 186.986558 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_198\">\n", "    <path d=\"M 445.956212 188.638125 \n", "L 447.937762 188.638125 \n", "L 447.937762 187.000239 \n", "L 445.956212 187.000239 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_199\">\n", "    <path d=\"M 447.937762 188.638125 \n", "L 449.919291 188.638125 \n", "L 449.919291 187.325622 \n", "L 447.937762 187.325622 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_200\">\n", "    <path d=\"M 449.919291 188.638125 \n", "L 451.90084 188.638125 \n", "L 451.90084 187.443206 \n", "L 449.919291 187.443206 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_201\">\n", "    <path d=\"M 451.90083 188.638125 \n", "L 453.88236 188.638125 \n", "L 453.88236 187.541638 \n", "L 451.90083 187.541638 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_202\">\n", "    <path d=\"M 453.88235 188.638125 \n", "L 455.863899 188.638125 \n", "L 455.863899 187.555315 \n", "L 453.88235 187.555315 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_203\">\n", "    <path d=\"M 455.863899 188.638125 \n", "L 457.845428 188.638125 \n", "L 457.845428 187.609997 \n", "L 455.863899 187.609997 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_204\">\n", "    <path d=\"M 457.845428 188.638125 \n", "L 459.826977 188.638125 \n", "L 459.826977 187.820549 \n", "L 457.845428 187.820549 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_205\">\n", "    <path d=\"M 459.826968 188.638125 \n", "L 461.808497 188.638125 \n", "L 461.808497 187.943592 \n", "L 459.826968 187.943592 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_206\">\n", "    <path d=\"M 461.808487 188.638125 \n", "L 463.790036 188.638125 \n", "L 463.790036 188.003752 \n", "L 461.808487 188.003752 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_207\">\n", "    <path d=\"M 463.790036 188.638125 \n", "L 465.771566 188.638125 \n", "L 465.771566 188.039295 \n", "L 463.790036 188.039295 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_208\">\n", "    <path d=\"M 465.771566 188.638125 \n", "L 467.753115 188.638125 \n", "L 467.753115 188.148673 \n", "L 465.771566 188.148673 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_209\">\n", "    <path d=\"M 467.753105 188.638125 \n", "L 469.734634 188.638125 \n", "L 469.734634 188.230702 \n", "L 467.753105 188.230702 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_210\">\n", "    <path d=\"M 469.734624 188.638125 \n", "L 471.716173 188.638125 \n", "L 471.716173 188.214298 \n", "L 469.734624 188.214298 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_211\">\n", "    <path d=\"M 471.716173 188.638125 \n", "L 473.697703 188.638125 \n", "L 473.697703 188.244374 \n", "L 471.716173 188.244374 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_212\">\n", "    <path d=\"M 473.697703 188.638125 \n", "L 475.679232 188.638125 \n", "L 475.679232 188.383827 \n", "L 473.697703 188.383827 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_213\">\n", "    <path d=\"M 475.679232 188.638125 \n", "L 477.660781 188.638125 \n", "L 477.660781 188.370157 \n", "L 475.679232 188.370157 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_214\">\n", "    <path d=\"M 477.660771 188.638125 \n", "L 479.642301 188.638125 \n", "L 479.642301 188.320937 \n", "L 477.660771 188.320937 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_215\">\n", "    <path d=\"M 479.642291 188.638125 \n", "L 481.62384 188.638125 \n", "L 481.62384 188.427579 \n", "L 479.642291 188.427579 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_216\">\n", "    <path d=\"M 481.62384 188.638125 \n", "L 483.605369 188.638125 \n", "L 483.605369 188.413906 \n", "L 481.62384 188.413906 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_217\">\n", "    <path d=\"M 483.60535 188.638125 \n", "L 485.586879 188.638125 \n", "L 485.586879 188.493203 \n", "L 483.60535 188.493203 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_218\">\n", "    <path d=\"M 485.586899 188.638125 \n", "L 487.568448 188.638125 \n", "L 487.568448 188.44672 \n", "L 485.586899 188.44672 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_219\">\n", "    <path d=\"M 487.568448 188.638125 \n", "L 489.549977 188.638125 \n", "L 489.549977 188.498672 \n", "L 487.568448 188.498672 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_220\">\n", "    <path d=\"M 489.549958 188.638125 \n", "L 491.531487 188.638125 \n", "L 491.531487 188.52875 \n", "L 489.549958 188.52875 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_221\">\n", "    <path d=\"M 491.531507 188.638125 \n", "L 493.513036 188.638125 \n", "L 493.513036 188.545156 \n", "L 491.531507 188.545156 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_222\">\n", "    <path d=\"M 493.513036 188.638125 \n", "L 495.494585 188.638125 \n", "L 495.494585 188.556094 \n", "L 493.513036 188.556094 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_223\">\n", "    <path d=\"M 495.494565 188.638125 \n", "L 497.476095 188.638125 \n", "L 497.476095 188.52875 \n", "L 495.494565 188.52875 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_224\">\n", "    <path d=\"M 497.476115 188.638125 \n", "L 499.457644 188.638125 \n", "L 499.457644 188.556094 \n", "L 497.476115 188.556094 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_225\">\n", "    <path d=\"M 499.457624 188.638125 \n", "L 501.439154 188.638125 \n", "L 501.439154 188.569765 \n", "L 499.457624 188.569765 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_226\">\n", "    <path d=\"M 501.439173 188.638125 \n", "L 503.420723 188.638125 \n", "L 503.420723 188.567032 \n", "L 501.439173 188.567032 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_227\">\n", "    <path d=\"M 503.420723 188.638125 \n", "L 505.402252 188.638125 \n", "L 505.402252 188.558828 \n", "L 503.420723 188.558828 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_228\">\n", "    <path d=\"M 505.402232 188.638125 \n", "L 507.383761 188.638125 \n", "L 507.383761 188.580703 \n", "L 505.402232 188.580703 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_229\">\n", "    <path d=\"M 507.383781 188.638125 \n", "L 509.365311 188.638125 \n", "L 509.365311 188.588906 \n", "L 507.383781 188.588906 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_230\">\n", "    <path d=\"M 509.365291 188.638125 \n", "L 511.34682 188.638125 \n", "L 511.34682 188.583437 \n", "L 509.365291 188.583437 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_231\">\n", "    <path d=\"M 511.34684 188.638125 \n", "L 513.328389 188.638125 \n", "L 513.328389 188.594375 \n", "L 511.34684 188.594375 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_232\">\n", "    <path d=\"M 513.328389 188.638125 \n", "L 515.309919 188.638125 \n", "L 515.309919 188.586172 \n", "L 513.328389 188.586172 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_233\">\n", "    <path d=\"M 515.309899 188.638125 \n", "L 517.291428 188.638125 \n", "L 517.291428 188.605312 \n", "L 515.309899 188.605312 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_234\">\n", "    <path d=\"M 517.291448 188.638125 \n", "L 519.272977 188.638125 \n", "L 519.272977 188.624453 \n", "L 517.291448 188.624453 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_235\">\n", "    <path d=\"M 519.272977 188.638125 \n", "L 521.254527 188.638125 \n", "L 521.254527 188.599844 \n", "L 519.272977 188.599844 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_236\">\n", "    <path d=\"M 521.254507 188.638125 \n", "L 523.236036 188.638125 \n", "L 523.236036 188.610781 \n", "L 521.254507 188.610781 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_237\">\n", "    <path d=\"M 523.236056 188.638125 \n", "L 525.217585 188.638125 \n", "L 525.217585 188.618984 \n", "L 523.236056 188.618984 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_238\">\n", "    <path d=\"M 525.217565 188.638125 \n", "L 527.199095 188.638125 \n", "L 527.199095 188.618984 \n", "L 525.217565 188.618984 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_239\">\n", "    <path d=\"M 527.199115 188.638125 \n", "L 529.180664 188.638125 \n", "L 529.180664 188.61625 \n", "L 527.199115 188.61625 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_240\">\n", "    <path d=\"M 529.180664 188.638125 \n", "L 531.162193 188.638125 \n", "L 531.162193 188.618984 \n", "L 529.180664 188.618984 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_241\">\n", "    <path d=\"M 531.162173 188.638125 \n", "L 533.143703 188.638125 \n", "L 533.143703 188.621719 \n", "L 531.162173 188.621719 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_242\">\n", "    <path d=\"M 533.143723 188.638125 \n", "L 535.125252 188.638125 \n", "L 535.125252 188.621719 \n", "L 533.143723 188.621719 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_243\">\n", "    <path d=\"M 535.125232 188.638125 \n", "L 537.106761 188.638125 \n", "L 537.106761 188.621719 \n", "L 535.125232 188.621719 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_244\">\n", "    <path d=\"M 537.106781 188.638125 \n", "L 539.088331 188.638125 \n", "L 539.088331 188.629922 \n", "L 537.106781 188.629922 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_245\">\n", "    <path d=\"M 539.088331 188.638125 \n", "L 541.06986 188.638125 \n", "L 541.06986 188.632656 \n", "L 539.088331 188.632656 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_246\">\n", "    <path d=\"M 541.06984 188.638125 \n", "L 543.051369 188.638125 \n", "L 543.051369 188.627187 \n", "L 541.06984 188.627187 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_247\">\n", "    <path d=\"M 543.051389 188.638125 \n", "L 545.032919 188.638125 \n", "L 545.032919 188.635391 \n", "L 543.051389 188.635391 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_248\">\n", "    <path d=\"M 545.032919 188.638125 \n", "L 547.014468 188.638125 \n", "L 547.014468 188.629922 \n", "L 545.032919 188.629922 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_249\">\n", "    <path d=\"M 547.014448 188.638125 \n", "L 548.995977 188.638125 \n", "L 548.995977 188.627187 \n", "L 547.014448 188.627187 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_250\">\n", "    <path d=\"M 548.995997 188.638125 \n", "L 550.977527 188.638125 \n", "L 550.977527 188.632656 \n", "L 548.995997 188.632656 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_251\">\n", "    <path d=\"M 550.977507 188.638125 \n", "L 552.959036 188.638125 \n", "L 552.959036 188.635391 \n", "L 550.977507 188.635391 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_252\">\n", "    <path d=\"M 552.959056 188.638125 \n", "L 554.940605 188.638125 \n", "L 554.940605 188.635391 \n", "L 552.959056 188.635391 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_253\">\n", "    <path d=\"M 554.940605 188.638125 \n", "L 556.922134 188.638125 \n", "L 556.922134 188.629922 \n", "L 554.940605 188.629922 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_254\">\n", "    <path d=\"M 556.922115 188.638125 \n", "L 558.903644 188.638125 \n", "L 558.903644 188.638125 \n", "L 556.922115 188.638125 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_255\">\n", "    <path d=\"M 558.903664 188.638125 \n", "L 560.885193 188.638125 \n", "L 560.885193 188.629922 \n", "L 558.903664 188.629922 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_256\">\n", "    <path d=\"M 560.885193 188.638125 \n", "L 562.866742 188.638125 \n", "L 562.866742 188.635391 \n", "L 560.885193 188.635391 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_257\">\n", "    <path d=\"M 562.866723 188.638125 \n", "L 564.848252 188.638125 \n", "L 564.848252 188.629922 \n", "L 562.866723 188.629922 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_258\">\n", "    <path d=\"M 564.848272 188.638125 \n", "L 566.829801 188.638125 \n", "L 566.829801 188.635391 \n", "L 564.848272 188.635391 \n", "z\n", "\" clip-path=\"url(#pa20a266b46)\" style=\"fill: #4c72b0; fill-opacity: 0.5; stroke: #4c72b0; stroke-linejoin: miter\"/>\n", "   </g>\n", "   <g id=\"patch_259\">\n", "    <path d=\"M 34.193438 188.638125 \n", "L 34.193438 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_260\">\n", "    <path d=\"M 592.193437 188.638125 \n", "L 592.193437 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_261\">\n", "    <path d=\"M 34.193438 188.638125 \n", "L 592.193438 188.638125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"patch_262\">\n", "    <path d=\"M 34.193438 22.318125 \n", "L 592.193438 22.318125 \n", "\" style=\"fill: none; stroke: #ffffff; stroke-width: 1.25; stroke-linejoin: miter; stroke-linecap: square\"/>\n", "   </g>\n", "   <g id=\"text_10\">\n", "    <!-- Variational dequantization -->\n", "    <g style=\"fill: #262626\" transform=\"translate(234.39375 16.318125)scale(0.12 -0.12)\">\n", "     <defs>\n", "      <path id=\"DejaVuSans-56\" d=\"M 1831 0 \n", "L 50 4666 \n", "L 709 4666 \n", "L 2188 738 \n", "L 3669 4666 \n", "L 4325 4666 \n", "L 2547 0 \n", "L 1831 0 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-61\" d=\"M 2194 1759 \n", "Q 1497 1759 1228 1600 \n", "Q 959 1441 959 1056 \n", "Q 959 750 1161 570 \n", "Q 1363 391 1709 391 \n", "Q 2188 391 2477 730 \n", "Q 2766 1069 2766 1631 \n", "L 2766 1759 \n", "L 2194 1759 \n", "z\n", "M 3341 1997 \n", "L 3341 0 \n", "L 2766 0 \n", "L 2766 531 \n", "Q 2569 213 2275 61 \n", "Q 1981 -91 1556 -91 \n", "Q 1019 -91 701 211 \n", "Q 384 513 384 1019 \n", "Q 384 1609 779 1909 \n", "Q 1175 2209 1959 2209 \n", "L 2766 2209 \n", "L 2766 2266 \n", "Q 2766 2663 2505 2880 \n", "Q 2244 3097 1772 3097 \n", "Q 1472 3097 1187 3025 \n", "Q 903 2953 641 2809 \n", "L 641 3341 \n", "Q 956 3463 1253 3523 \n", "Q 1550 3584 1831 3584 \n", "Q 2591 3584 2966 3190 \n", "Q 3341 2797 3341 1997 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-72\" d=\"M 2631 2963 \n", "Q 2534 3019 2420 3045 \n", "Q 2306 3072 2169 3072 \n", "Q 1681 3072 1420 2755 \n", "Q 1159 2438 1159 1844 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 3500 \n", "L 1159 3500 \n", "L 1159 2956 \n", "Q 1341 3275 1631 3429 \n", "Q 1922 3584 2338 3584 \n", "Q 2397 3584 2469 3576 \n", "Q 2541 3569 2628 3553 \n", "L 2631 2963 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-69\" d=\"M 603 3500 \n", "L 1178 3500 \n", "L 1178 0 \n", "L 603 0 \n", "L 603 3500 \n", "z\n", "M 603 4863 \n", "L 1178 4863 \n", "L 1178 4134 \n", "L 603 4134 \n", "L 603 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-74\" d=\"M 1172 4494 \n", "L 1172 3500 \n", "L 2356 3500 \n", "L 2356 3053 \n", "L 1172 3053 \n", "L 1172 1153 \n", "Q 1172 725 1289 603 \n", "Q 1406 481 1766 481 \n", "L 2356 481 \n", "L 2356 0 \n", "L 1766 0 \n", "Q 1100 0 847 248 \n", "Q 594 497 594 1153 \n", "L 594 3053 \n", "L 172 3053 \n", "L 172 3500 \n", "L 594 3500 \n", "L 594 4494 \n", "L 1172 4494 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6f\" d=\"M 1959 3097 \n", "Q 1497 3097 1228 2736 \n", "Q 959 2375 959 1747 \n", "Q 959 1119 1226 758 \n", "Q 1494 397 1959 397 \n", "Q 2419 397 2687 759 \n", "Q 2956 1122 2956 1747 \n", "Q 2956 2369 2687 2733 \n", "Q 2419 3097 1959 3097 \n", "z\n", "M 1959 3584 \n", "Q 2709 3584 3137 3096 \n", "Q 3566 2609 3566 1747 \n", "Q 3566 888 3137 398 \n", "Q 2709 -91 1959 -91 \n", "Q 1206 -91 779 398 \n", "Q 353 888 353 1747 \n", "Q 353 2609 779 3096 \n", "Q 1206 3584 1959 3584 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6e\" d=\"M 3513 2113 \n", "L 3513 0 \n", "L 2938 0 \n", "L 2938 2094 \n", "Q 2938 2591 2744 2837 \n", "Q 2550 3084 2163 3084 \n", "Q 1697 3084 1428 2787 \n", "Q 1159 2491 1159 1978 \n", "L 1159 0 \n", "L 581 0 \n", "L 581 3500 \n", "L 1159 3500 \n", "L 1159 2956 \n", "Q 1366 3272 1645 3428 \n", "Q 1925 3584 2291 3584 \n", "Q 2894 3584 3203 3211 \n", "Q 3513 2838 3513 2113 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-6c\" d=\"M 603 4863 \n", "L 1178 4863 \n", "L 1178 0 \n", "L 603 0 \n", "L 603 4863 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-20\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-64\" d=\"M 2906 2969 \n", "L 2906 4863 \n", "L 3481 4863 \n", "L 3481 0 \n", "L 2906 0 \n", "L 2906 525 \n", "Q 2725 213 2448 61 \n", "Q 2172 -91 1784 -91 \n", "Q 1150 -91 751 415 \n", "Q 353 922 353 1747 \n", "Q 353 2572 751 3078 \n", "Q 1150 3584 1784 3584 \n", "Q 2172 3584 2448 3432 \n", "Q 2725 3281 2906 2969 \n", "z\n", "M 947 1747 \n", "Q 947 1113 1208 752 \n", "Q 1469 391 1925 391 \n", "Q 2381 391 2643 752 \n", "Q 2906 1113 2906 1747 \n", "Q 2906 2381 2643 2742 \n", "Q 2381 3103 1925 3103 \n", "Q 1469 3103 1208 2742 \n", "Q 947 2381 947 1747 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-65\" d=\"M 3597 1894 \n", "L 3597 1613 \n", "L 953 1613 \n", "Q 991 1019 1311 708 \n", "Q 1631 397 2203 397 \n", "Q 2534 397 2845 478 \n", "Q 3156 559 3463 722 \n", "L 3463 178 \n", "Q 3153 47 2828 -22 \n", "Q 2503 -91 2169 -91 \n", "Q 1331 -91 842 396 \n", "Q 353 884 353 1716 \n", "Q 353 2575 817 3079 \n", "Q 1281 3584 2069 3584 \n", "Q 2775 3584 3186 3129 \n", "Q 3597 2675 3597 1894 \n", "z\n", "M 3022 2063 \n", "Q 3016 2534 2758 2815 \n", "Q 2500 3097 2075 3097 \n", "Q 1594 3097 1305 2825 \n", "Q 1016 2553 972 2059 \n", "L 3022 2063 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-71\" d=\"M 947 1747 \n", "Q 947 1113 1208 752 \n", "Q 1469 391 1925 391 \n", "Q 2381 391 2643 752 \n", "Q 2906 1113 2906 1747 \n", "Q 2906 2381 2643 2742 \n", "Q 2381 3103 1925 3103 \n", "Q 1469 3103 1208 2742 \n", "Q 947 2381 947 1747 \n", "z\n", "M 2906 525 \n", "Q 2725 213 2448 61 \n", "Q 2172 -91 1784 -91 \n", "Q 1150 -91 751 415 \n", "Q 353 922 353 1747 \n", "Q 353 2572 751 3078 \n", "Q 1150 3584 1784 3584 \n", "Q 2172 3584 2448 3432 \n", "Q 2725 3281 2906 2969 \n", "L 2906 3500 \n", "L 3481 3500 \n", "L 3481 -1331 \n", "L 2906 -1331 \n", "L 2906 525 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-75\" d=\"M 544 1381 \n", "L 544 3500 \n", "L 1119 3500 \n", "L 1119 1403 \n", "Q 1119 906 1312 657 \n", "Q 1506 409 1894 409 \n", "Q 2359 409 2629 706 \n", "Q 2900 1003 2900 1516 \n", "L 2900 3500 \n", "L 3475 3500 \n", "L 3475 0 \n", "L 2900 0 \n", "L 2900 538 \n", "Q 2691 219 2414 64 \n", "Q 2138 -91 1772 -91 \n", "Q 1169 -91 856 284 \n", "Q 544 659 544 1381 \n", "z\n", "M 1991 3584 \n", "L 1991 3584 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "      <path id=\"DejaVuSans-7a\" d=\"M 353 3500 \n", "L 3084 3500 \n", "L 3084 2975 \n", "L 922 459 \n", "L 3084 459 \n", "L 3084 0 \n", "L 275 0 \n", "L 275 525 \n", "L 2438 3041 \n", "L 353 3041 \n", "L 353 3500 \n", "z\n", "\" transform=\"scale(0.015625)\"/>\n", "     </defs>\n", "     <use xlink:href=\"#DejaVuSans-56\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"60.658203\"/>\n", "     <use xlink:href=\"#DejaVuSans-72\" x=\"121.9375\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"163.050781\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"190.833984\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"252.113281\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"291.322266\"/>\n", "     <use xlink:href=\"#DejaVuSans-6f\" x=\"319.105469\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"380.287109\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"443.666016\"/>\n", "     <use xlink:href=\"#DejaVuSans-6c\" x=\"504.945312\"/>\n", "     <use xlink:href=\"#DejaVuSans-20\" x=\"532.728516\"/>\n", "     <use xlink:href=\"#DejaVuSans-64\" x=\"564.515625\"/>\n", "     <use xlink:href=\"#DejaVuSans-65\" x=\"627.992188\"/>\n", "     <use xlink:href=\"#DejaVuSans-71\" x=\"689.515625\"/>\n", "     <use xlink:href=\"#DejaVuSans-75\" x=\"752.992188\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"816.371094\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"877.650391\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"941.029297\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"980.238281\"/>\n", "     <use xlink:href=\"#DejaVuSans-7a\" x=\"1008.021484\"/>\n", "     <use xlink:href=\"#DejaVuSans-61\" x=\"1060.511719\"/>\n", "     <use xlink:href=\"#DejaVuSans-74\" x=\"1121.791016\"/>\n", "     <use xlink:href=\"#DejaVuSans-69\" x=\"1161\"/>\n", "     <use xlink:href=\"#DejaVuSans-6f\" x=\"1188.783203\"/>\n", "     <use xlink:href=\"#DejaVuSans-6e\" x=\"1249.964844\"/>\n", "    </g>\n", "   </g>\n", "  </g>\n", " </g>\n", " <defs>\n", "  <clipPath id=\"pa20a266b46\">\n", "   <rect x=\"34.193438\" y=\"22.318125\" width=\"558\" height=\"166.32\"/>\n", "  </clipPath>\n", " </defs>\n", "</svg>\n"], "text/plain": ["<Figure size 1000x300 with 1 Axes>"]}, "metadata": {}, "output_type": "display_data"}], "source": ["visualize_dequant_distribution(flow_dict[\"vardeq\"][\"model\"], sample_imgs, title=\"Variational dequantization\")"]}, {"cell_type": "markdown", "id": "7c10c6d8", "metadata": {"papermill": {"duration": 0.029624, "end_time": "2022-05-12T12:19:21.267494", "exception": false, "start_time": "2022-05-12T12:19:21.237870", "status": "completed"}, "tags": []}, "source": ["The dequantization distribution in the first plot shows that the MNIST images have a strong bias towards 0 (black),\n", "and the distribution of them have a sharp border as mentioned before.\n", "The variational dequantization module has indeed learned a much smoother distribution with a Gaussian-like curve which can be modeled much better.\n", "For the other values, we would need to visualize the distribution $q(u|x)$ on a deeper level, depending on $x$.\n", "However, as all $u$'s interact and depend on each other, we would need\n", "to visualize a distribution in 784 dimensions, which is not that\n", "intuitive anymore."]}, {"cell_type": "markdown", "id": "620279fc", "metadata": {"papermill": {"duration": 0.029553, "end_time": "2022-05-12T12:19:21.325422", "exception": false, "start_time": "2022-05-12T12:19:21.295869", "status": "completed"}, "tags": []}, "source": ["## Conclusion\n", "\n", "In conclusion, we have seen how to implement our own normalizing flow, and what difficulties arise if we want to apply them on images.\n", "Dequantization is a crucial step in mapping the discrete images into continuous space to prevent underisable delta-peak solutions.\n", "While dequantization creates hypercubes with hard border, variational dequantization allows us to fit a flow much better on the data.\n", "This allows us to obtain a lower bits per dimension score, while not affecting the sampling speed.\n", "The most common flow element, the coupling layer, is simple to implement, and yet effective.\n", "Furthermore, multi-scale architectures help to capture the global image context while allowing us to efficiently scale up the flow.\n", "Normalizing flows are an interesting alternative to VAEs as they allow an exact likelihood estimate in continuous space,\n", "and we have the guarantee that every possible input $x$ has a corresponding latent vector $z$.\n", "However, even beyond continuous inputs and images, flows can be applied and allow us to exploit\n", "the data structure in latent space, as e.g. on graphs for the task of molecule generation [6].\n", "Recent advances in [Neural ODEs](https://arxiv.org/pdf/1806.07366.pdf) allow a flow with infinite number of layers,\n", "called Continuous Normalizing Flows, whose potential is yet to fully explore.\n", "Overall, normalizing flows are an exciting research area which will continue over the next couple of years."]}, {"cell_type": "markdown", "id": "27e38410", "metadata": {"papermill": {"duration": 0.029009, "end_time": "2022-05-12T12:19:21.385707", "exception": false, "start_time": "2022-05-12T12:19:21.356698", "status": "completed"}, "tags": []}, "source": ["## References\n", "\n", "[1] Dinh, L., Sohl-Dickstein, J., and Bengio, S. (2017).\n", "\u201cDensity estimation using Real NVP,\u201d In: 5th International Conference on Learning Representations, ICLR 2017.\n", "[Link](https://arxiv.org/abs/1605.08803)\n", "\n", "[2] Kingma, D. P., and Dhariwal, P. (2018).\n", "\u201cGlow: Generative Flow with Invertible 1x1 Convolutions,\u201d In: Advances in Neural Information Processing Systems, vol.\n", "31, pp.\n", "10215--10224.\n", "[Link](http://papers.nips.cc/paper/8224-glow-generative-flow-with-invertible-1x1-convolutions.pdf)\n", "\n", "[3] Ho, J., Chen, X., Srinivas, A., Duan, Y., and Abbeel, P. (2019).\n", "\u201cFlow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design,\u201d\n", "in Proceedings of the 36th International Conference on Machine Learning, vol.\n", "97, pp.\n", "2722\u20132730.\n", "[Link](https://arxiv.org/abs/1902.00275)\n", "\n", "[4] Durkan, C., Bekasov, A., Murray, I., and Papamakarios, G. (2019).\n", "\u201cNeural Spline Flows,\u201d In: Advances in Neural Information Processing Systems, pp.\n", "7509\u20137520.\n", "[Link](http://papers.neurips.cc/paper/8969-neural-spline-flows.pdf)\n", "\n", "[5] Hoogeboom, E., Cohen, T. S., and Tomczak, J. M. (2020).\n", "\u201cLearning Discrete Distributions by Dequantization,\u201d arXiv preprint arXiv2001.11235v1.\n", "[Link](https://arxiv.org/abs/2001.11235)\n", "\n", "[6] Lippe, P., and Gavves, E. (2021).\n", "\u201cCategorical Normalizing Flows via Continuous Transformations,\u201d\n", "In: International Conference on Learning Representations, ICLR 2021.\n", "[Link](https://openreview.net/pdf?id=-GLNZeVDuik)"]}, {"cell_type": "markdown", "id": "48804717", "metadata": {"papermill": {"duration": 0.029953, "end_time": "2022-05-12T12:19:21.444776", "exception": false, "start_time": "2022-05-12T12:19:21.414823", "status": "completed"}, "tags": []}, "source": ["## Congratulations - Time to Join the Community!\n", "\n", "Congratulations on completing this notebook tutorial! If you enjoyed this and would like to join the Lightning\n", "movement, you can do so in the following ways!\n", "\n", "### Star [Lightning](https://github.com/PyTorchLightning/pytorch-lightning) on GitHub\n", "The easiest way to help our community is just by starring the GitHub repos! This helps raise awareness of the cool\n", "tools we're building.\n", "\n", "### Join our [Slack](https://www.pytorchlightning.ai/community)!\n", "The best way to keep up to date on the latest advancements is to join our community! Make sure to introduce yourself\n", "and share your interests in `#general` channel\n", "\n", "\n", "### Contributions !\n", "The best way to contribute to our community is to become a code contributor! At any time you can go to\n", "[Lightning](https://github.com/PyTorchLightning/pytorch-lightning) or [Bolt](https://github.com/PyTorchLightning/lightning-bolts)\n", "GitHub Issues page and filter for \"good first issue\".\n", "\n", "* [Lightning good first issue](https://github.com/PyTorchLightning/pytorch-lightning/issues?q=is%3Aopen+is%3Aissue+label%3A%22good+first+issue%22)\n", "* [Bolt good first issue](https://github.com/PyTorchLightning/lightning-bolts/issues?q=is%3Aopen+is%3Aissue+label%3A%22good+first+issue%22)\n", "* You can also contribute your own notebooks with useful examples !\n", "\n", "### Great thanks from the entire Pytorch Lightning Team for your interest !\n", "\n", "[![Pytorch Lightning](data:image/png;base64,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){height=\"60px\" width=\"240px\"}](https://pytorchlightning.ai)"]}, {"cell_type": "raw", "metadata": {"raw_mimetype": "text/restructuredtext"}, "source": [".. customcarditem::\n", "   :header: Tutorial 9: Normalizing Flows for Image Modeling\n", "   :card_description: In this tutorial, we will take a closer look at complex, deep normalizing flows. The most popular, current application of deep normalizing flows is to model datasets of...\n", "   :tags: Image,GPU/TPU,UvA-DL-Course\n", "   :image: _static/images/course_UvA-DL/09-normalizing-flows.jpg"]}], "metadata": {"jupytext": {"cell_metadata_filter": "colab_type,id,colab,-all", "formats": "ipynb,py:percent", "main_language": "python"}, "language_info": {"codemirror_mode": {"name": "ipython", "version": 3}, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.8.10"}, "papermill": {"default_parameters": {}, "duration": 25.941431, "end_time": "2022-05-12T12:19:23.098784", "environment_variables": {}, "exception": null, "input_path": "course_UvA-DL/09-normalizing-flows/NF_image_modeling.ipynb", "output_path": ".notebooks/course_UvA-DL/09-normalizing-flows.ipynb", "parameters": {}, "start_time": "2022-05-12T12:18:57.157353", "version": "2.3.4"}, "widgets": {"application/vnd.jupyter.widget-state+json": {"state": {"033751fff3854e7fba6e527fc3ca1af0": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "0a53e8d36171453087899d7b05d0dab4": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "0bf513f5ede34bf0aa6095e88896ee01": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HTMLModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HTMLView", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_f26ee6645b204d699c8d6a262817a8ef", "placeholder": "\u200b", "style": "IPY_MODEL_101cb19c084044df8919210038562038", "value": ""}}, "101cb19c084044df8919210038562038": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "DescriptionStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "DescriptionStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "description_width": ""}}, "111dcbf1598644c6b5fe894526a16bc9": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "1474b267cbb442cf939e8c460db28ed9": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "ProgressStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "ProgressStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "bar_color": null, "description_width": ""}}, "1df07eff5dec46f7b8867701f6c9b5f8": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "FloatProgressModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "FloatProgressModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "ProgressView", "bar_style": "success", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_abdfdb58fdeb414cb9bd9b21fcd9fa99", "max": 4542.0, "min": 0.0, "orientation": "horizontal", "style": "IPY_MODEL_1474b267cbb442cf939e8c460db28ed9", "value": 4542.0}}, "218074dccea7473e816a8461d3bec333": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "ProgressStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "ProgressStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "bar_color": null, "description_width": ""}}, "22731f4074314260bf3a8474c377ec03": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HTMLModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HTMLView", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_8b8edc12175c48719a46a007eb575ec4", "placeholder": "\u200b", "style": "IPY_MODEL_4edbc364f3154c35b86cec4dfd3fe383", "value": " 62%"}}, "245169ab03414848bc71b05cf2129307": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "ProgressStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "ProgressStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "bar_color": null, "description_width": ""}}, "26c84ae6074a49cd95b93cd701457f0d": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HTMLModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HTMLView", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_0a53e8d36171453087899d7b05d0dab4", "placeholder": "\u200b", "style": "IPY_MODEL_a1dfc3dd292a4c3f9a440d34dd2bafa9", "value": "  0%"}}, "2739cef7a26c4d36abe27140463f1536": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "DescriptionStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "DescriptionStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "description_width": ""}}, "280b89d79e1f40bab2d9ca74d447480d": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "DescriptionStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "DescriptionStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "description_width": ""}}, "2b08b3d74913422d99ad29d9b9b39904": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "DescriptionStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "DescriptionStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "description_width": ""}}, "2b2f05b4a3d44430b7343f7c7dd96d6b": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "2e0f9562244644ccac867ccd8b85d15f": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "FloatProgressModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "FloatProgressModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "ProgressView", "bar_style": "success", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_9bd7ad04b5bc4639bdd0eef00c52e3a9", "max": 9912422.0, "min": 0.0, "orientation": "horizontal", "style": "IPY_MODEL_9dac662d19214eddaae33c0e40a93cef", "value": 9912422.0}}, "316381f85ead4a8d865c45dc0aaf4c74": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "432fa993cc574f50963c5a558841af95": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "DescriptionStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "DescriptionStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "description_width": ""}}, "44514b44ec4f42baba6c3b8a0a05cf69": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HTMLModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HTMLView", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_316381f85ead4a8d865c45dc0aaf4c74", "placeholder": "\u200b", "style": "IPY_MODEL_280b89d79e1f40bab2d9ca74d447480d", "value": " 1649664/? [00:00&lt;00:00, 48914787.24it/s]"}}, "453d7b4525594a01a1023e7484e02c8c": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "DescriptionStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "DescriptionStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "description_width": ""}}, "4762650b38ca460ebf7a8b7eaf2d2125": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "4edbc364f3154c35b86cec4dfd3fe383": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "DescriptionStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "DescriptionStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "description_width": ""}}, "560c48ca02ef41c09de8dc7078c4bf24": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "5a571cfc40ac454b8583b4b978c5f5a8": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HTMLModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HTMLView", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_b845e4639c67471f86e22528305643ca", "placeholder": "\u200b", "style": "IPY_MODEL_e8041c5ef7e74a60b5527813002dcc5b", "value": ""}}, "5b1adf06f69f4853b5475a5eb0b8ed55": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "5cc75509b1e64556a1358f725e34afd0": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "FloatProgressModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "FloatProgressModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "ProgressView", "bar_style": "", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_4762650b38ca460ebf7a8b7eaf2d2125", "max": 8.0, "min": 0.0, "orientation": "horizontal", "style": "IPY_MODEL_245169ab03414848bc71b05cf2129307", "value": 8.0}}, "5cd28bfb90d94e71b16c3cba25520349": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "FloatProgressModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "FloatProgressModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "ProgressView", "bar_style": "success", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_ca3de77b07124ce386d85a29236f3178", "max": 28881.0, "min": 0.0, "orientation": "horizontal", "style": "IPY_MODEL_218074dccea7473e816a8461d3bec333", "value": 28881.0}}, "5e798f310d4b42208f60488f4b25167b": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "5ec66f1159e941a0b8a03d507ab59716": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HTMLModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HTMLView", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_111dcbf1598644c6b5fe894526a16bc9", "placeholder": "\u200b", "style": "IPY_MODEL_2739cef7a26c4d36abe27140463f1536", "value": " 5120/? [00:00&lt;00:00, 253232.04it/s]"}}, "6512366631a949678ad0cfd766ffba0a": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HBoxModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HBoxModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HBoxView", "box_style": "", "children": ["IPY_MODEL_8b23d7d073dc4fadb963d7611e5bdfda", "IPY_MODEL_2e0f9562244644ccac867ccd8b85d15f", "IPY_MODEL_8e7de690e370429aadb271d44c5dca88"], "layout": "IPY_MODEL_e0dec845c63045a59b58b11cb729b68a"}}, "75ad285edbad41fe83fdf09c4c8cf6c0": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HTMLModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HTMLView", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_8e0c54ee985f4ea3820c13e8b58a3f2d", "placeholder": "\u200b", "style": "IPY_MODEL_453d7b4525594a01a1023e7484e02c8c", "value": ""}}, "75f3db6cb8e14d1296d489ca26b01f24": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HBoxModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HBoxModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HBoxView", "box_style": "", "children": ["IPY_MODEL_0bf513f5ede34bf0aa6095e88896ee01", "IPY_MODEL_1df07eff5dec46f7b8867701f6c9b5f8", "IPY_MODEL_5ec66f1159e941a0b8a03d507ab59716"], "layout": "IPY_MODEL_ed18b5e93d8745a68a5041277a459d9d"}}, "7928f983aa4344859749c3737df8c87e": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "80a80e415bb24e459eef8028675f66ad": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HBoxModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HBoxModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HBoxView", "box_style": "", "children": ["IPY_MODEL_5a571cfc40ac454b8583b4b978c5f5a8", "IPY_MODEL_f77de1e20aa44f558370b6ab4ab63b65", "IPY_MODEL_44514b44ec4f42baba6c3b8a0a05cf69"], "layout": "IPY_MODEL_2b2f05b4a3d44430b7343f7c7dd96d6b"}}, "8a5581805ad545c6a4cc26e86a64d636": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "DescriptionStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "DescriptionStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "description_width": ""}}, "8afbd845be84491384ba3254f348da55": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "8b23d7d073dc4fadb963d7611e5bdfda": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HTMLModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HTMLView", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_91acf3221c8a4ac2b3468f63e45358ae", "placeholder": "\u200b", "style": "IPY_MODEL_8a5581805ad545c6a4cc26e86a64d636", "value": ""}}, "8b8edc12175c48719a46a007eb575ec4": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "8e0c54ee985f4ea3820c13e8b58a3f2d": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "8e7de690e370429aadb271d44c5dca88": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HTMLModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HTMLView", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_8afbd845be84491384ba3254f348da55", "placeholder": "\u200b", "style": "IPY_MODEL_2b08b3d74913422d99ad29d9b9b39904", "value": " 9913344/? [00:00&lt;00:00, 102366614.04it/s]"}}, "91acf3221c8a4ac2b3468f63e45358ae": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "91eed8e58a8341e9aadfd56c1abbf3c1": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "94d70f1ca8af4a318c07f34a8ad16a23": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HTMLModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HTMLView", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_ecd7d7859e40414dad266aff3cf9908a", "placeholder": "\u200b", "style": "IPY_MODEL_432fa993cc574f50963c5a558841af95", "value": " 0/8 [00:00&lt;?, ?it/s]"}}, "9764df840c324bd79369088ec1295004": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "FloatProgressModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "FloatProgressModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "ProgressView", "bar_style": "", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_7928f983aa4344859749c3737df8c87e", "max": 8.0, "min": 0.0, "orientation": "horizontal", "style": "IPY_MODEL_c3b21e9f688e4d9fb9abb4b2bd35520c", "value": 8.0}}, "9a751f24d3c04ef7b8e5f4763ceccf4e": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "ProgressStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "ProgressStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "bar_color": null, "description_width": ""}}, "9bd7ad04b5bc4639bdd0eef00c52e3a9": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "9da52b7df5e8458e9237d8e64f99813c": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HTMLModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HTMLView", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_acfcf1a6a77a4f84ba076fa320e440f5", "placeholder": "\u200b", "style": "IPY_MODEL_a46709408d804941a191d4f4e94cdaff", "value": " 29696/? [00:00&lt;00:00, 1513047.27it/s]"}}, "9dac662d19214eddaae33c0e40a93cef": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "ProgressStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "ProgressStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "bar_color": null, "description_width": ""}}, "a05e28af0506439f8843dceab189d0a8": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HBoxModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HBoxModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HBoxView", "box_style": "", "children": ["IPY_MODEL_26c84ae6074a49cd95b93cd701457f0d", "IPY_MODEL_9764df840c324bd79369088ec1295004", "IPY_MODEL_94d70f1ca8af4a318c07f34a8ad16a23"], "layout": "IPY_MODEL_033751fff3854e7fba6e527fc3ca1af0"}}, "a1dfc3dd292a4c3f9a440d34dd2bafa9": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "DescriptionStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "DescriptionStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "description_width": ""}}, "a46709408d804941a191d4f4e94cdaff": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "DescriptionStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "DescriptionStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "description_width": ""}}, "abdfdb58fdeb414cb9bd9b21fcd9fa99": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "acfcf1a6a77a4f84ba076fa320e440f5": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "b845e4639c67471f86e22528305643ca": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "c3b21e9f688e4d9fb9abb4b2bd35520c": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "ProgressStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "ProgressStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "bar_color": null, "description_width": ""}}, "c49cd5fd7f434bd1bc57f2ce1bada81a": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HTMLModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HTMLModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HTMLView", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_91eed8e58a8341e9aadfd56c1abbf3c1", "placeholder": "\u200b", "style": "IPY_MODEL_e6cb260f2d2e4d9598aa26eded04b62c", "value": " 5/8 [00:00&lt;00:00, 49.34it/s]"}}, "ca3de77b07124ce386d85a29236f3178": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "ce6bcc8f63c346b4a80e0c245eb89704": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HBoxModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HBoxModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HBoxView", "box_style": "", "children": ["IPY_MODEL_75ad285edbad41fe83fdf09c4c8cf6c0", "IPY_MODEL_5cd28bfb90d94e71b16c3cba25520349", "IPY_MODEL_9da52b7df5e8458e9237d8e64f99813c"], "layout": "IPY_MODEL_5e798f310d4b42208f60488f4b25167b"}}, "d60486171a554fbfb03474a308153d0d": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "HBoxModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "HBoxModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "HBoxView", "box_style": "", "children": ["IPY_MODEL_22731f4074314260bf3a8474c377ec03", "IPY_MODEL_5cc75509b1e64556a1358f725e34afd0", "IPY_MODEL_c49cd5fd7f434bd1bc57f2ce1bada81a"], "layout": "IPY_MODEL_560c48ca02ef41c09de8dc7078c4bf24"}}, "e0dec845c63045a59b58b11cb729b68a": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "e6cb260f2d2e4d9598aa26eded04b62c": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "DescriptionStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "DescriptionStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "description_width": ""}}, "e8041c5ef7e74a60b5527813002dcc5b": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "DescriptionStyleModel", "state": {"_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "DescriptionStyleModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "StyleView", "description_width": ""}}, "ecd7d7859e40414dad266aff3cf9908a": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "ed18b5e93d8745a68a5041277a459d9d": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "f26ee6645b204d699c8d6a262817a8ef": {"model_module": "@jupyter-widgets/base", "model_module_version": "1.2.0", "model_name": "LayoutModel", "state": {"_model_module": "@jupyter-widgets/base", "_model_module_version": "1.2.0", "_model_name": "LayoutModel", "_view_count": null, "_view_module": "@jupyter-widgets/base", "_view_module_version": "1.2.0", "_view_name": "LayoutView", "align_content": null, "align_items": null, "align_self": null, "border": null, "bottom": null, "display": null, "flex": null, "flex_flow": null, "grid_area": null, "grid_auto_columns": null, "grid_auto_flow": null, "grid_auto_rows": null, "grid_column": null, "grid_gap": null, "grid_row": null, "grid_template_areas": null, "grid_template_columns": null, "grid_template_rows": null, "height": null, "justify_content": null, "justify_items": null, "left": null, "margin": null, "max_height": null, "max_width": null, "min_height": null, "min_width": null, "object_fit": null, "object_position": null, "order": null, "overflow": null, "overflow_x": null, "overflow_y": null, "padding": null, "right": null, "top": null, "visibility": null, "width": null}}, "f77de1e20aa44f558370b6ab4ab63b65": {"model_module": "@jupyter-widgets/controls", "model_module_version": "1.5.0", "model_name": "FloatProgressModel", "state": {"_dom_classes": [], "_model_module": "@jupyter-widgets/controls", "_model_module_version": "1.5.0", "_model_name": "FloatProgressModel", "_view_count": null, "_view_module": "@jupyter-widgets/controls", "_view_module_version": "1.5.0", "_view_name": "ProgressView", "bar_style": "success", "description": "", "description_tooltip": null, "layout": "IPY_MODEL_5b1adf06f69f4853b5475a5eb0b8ed55", "max": 1648877.0, "min": 0.0, "orientation": "horizontal", "style": "IPY_MODEL_9a751f24d3c04ef7b8e5f4763ceccf4e", "value": 1648877.0}}}, "version_major": 2, "version_minor": 0}}}, "nbformat": 4, "nbformat_minor": 5}