:orphan:

TorchMetrics' gallery
=====================

Welcome to a comprehensive guide on leveraging TorchMetrics, that facilitates the precise and consistent evaluation of machine learning models. As an integral tool for developers and researchers, TorchMetrics offers an array of metrics critical for assessing model performance across a variety of applications. Whether you are fine-tuning a neural network, comparing model iterations, or tracking performance improvements, this page provides a gallery of real-world applications where Torch Metrics can be effectively implemented.

By touring through this application gallery, users can gain insights into how TorchMetrics is utilized across different sectors and scale settings, empowering them with the knowledge to implement metrics effectively in their own projects. Whether your interest lies in academic research or commercial product development, the examples provided here will help demonstrate the versatility and utility of Torch Metrics in enhancing machine learning model assessment.



.. raw:: html

    <div class="sphx-glr-thumbnails">

.. thumbnail-parent-div-open

.. thumbnail-parent-div-close

.. raw:: html

    </div>

Audio domain
============

Audio-domain metrics are essential for assessing the performance of models in tasks such as speech recognition, audio classification, and sound event detection. TorchMetrics offers a comprehensive set of specialized metrics tailored for these audio-specific purposes. Utilizing these metrics from TorchMetrics aids in the development of more accurate and resilient audio-based models, ensuring that performance evaluations are both meaningful and directly applicable to real-world audio tasks.



.. raw:: html

    <div class="sphx-glr-thumbnails">

.. thumbnail-parent-div-open

.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This notebook will guide you through calculating the Perceptual Evaluation of Speech Quality (PESQ) score,  a key metric in assessing how effective noise reduction and enhancement techniques are in improving speech quality.  PESQ is widely adopted in industries such as telecommunications, VoIP, and audio processing.  It provides an objective way to measure the perceived quality of speech signals from a human listener&#x27;s perspective.">

.. only:: html

  .. image:: /gallery/audio/images/thumb/sphx_glr_pesq_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_audio_pesq.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Evaluating Speech Quality with PESQ metric</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Imagine developing a song recognition application. The software&#x27;s goal is to recognize a song even when it&#x27;s played in a noisy environment, similar to Shazam. To achieve this, you want to enhance the audio quality by reducing the noise and evaluating the improvement using the Signal-to-Noise Ratio (SNR).">

.. only:: html

  .. image:: /gallery/audio/images/thumb/sphx_glr_signal_to_noise_ratio_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_audio_signal_to_noise_ratio.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Signal-to-Noise Ratio</div>
    </div>


.. thumbnail-parent-div-close

.. raw:: html

    </div>

Image domain
============

Image-domain metrics are pivotal for gauging the performance of models in tasks like object detection, and segmentation. TorchMetrics provides a suite of specialized metrics designed for these purposes. Using these image-specific metrics from Torch Metrics helps in developing more precise and robust image-based models, ensuring that performance evaluations are both meaningful and directly applicable to practical vision tasks.



.. raw:: html

    <div class="sphx-glr-thumbnails">

.. thumbnail-parent-div-open

.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The Spatial Correlation Coefficient can be applied to compare the spatial structure of two images, which can be valuable in various domains such as medical imaging, remote sensing, and quality assessment in manufacturing or design processes.">

.. only:: html

  .. image:: /gallery/image/images/thumb/sphx_glr_spatial_correlation_coef_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_image_spatial_correlation_coef.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Spatial Correlation Coefficient</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The CLIPScore is a model-based image captioning metric that correlates well with human judgments.">

.. only:: html

  .. image:: /gallery/image/images/thumb/sphx_glr_clip_score_thumb.gif
    :alt:

  :ref:`sphx_glr_gallery_image_clip_score.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">CLIPScore</div>
    </div>


.. thumbnail-parent-div-close

.. raw:: html

    </div>

Text domain
============

Text-domain metrics are essential for evaluating the performance of models in tasks like text classification, summarization, and translation. TorchMetrics provides a suite of specialized metrics designed for these purposes. Using these text-specific metrics from Torch Metrics helps in developing more precise and robust text-based models, ensuring that performance evaluations are both meaningful and directly applicable to practical NLP tasks.



.. raw:: html

    <div class="sphx-glr-thumbnails">

.. thumbnail-parent-div-open

.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Perplexity is a measure of how well a probabilistic model predicts a sample.">

.. only:: html

  .. image:: /gallery/text/images/thumb/sphx_glr_perplexity_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_text_perplexity.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Perplexity</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="BERTScore is a text generation metric to compute the similarity between a generated text and a reference text using a pre-trained BERT model. Instead of relying on exact token matches, BERTScore leverages contextual embeddings to capture the semantic similarity between the texts. This makes BERTScore robust to paraphrasing and word order variations. BERTScore has been shown to correlate well with human judgments and is widely used in evaluating text generation models.">

.. only:: html

  .. image:: /gallery/text/images/thumb/sphx_glr_bertscore_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_text_bertscore.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">BERTScore</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="The ROUGE (Recall-Oriented Understudy for Gisting Evaluation) metric used to evaluate the quality of generated text compared to a reference text. It does so by computing the overlap between two texts, for which a subsequent precision and recall value can be computed. The ROUGE score is often used in the context of generative tasks such as text summarization and machine translation.">

.. only:: html

  .. image:: /gallery/text/images/thumb/sphx_glr_rouge_thumb.png
    :alt:

  :ref:`sphx_glr_gallery_text_rouge.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">ROUGE</div>
    </div>


.. thumbnail-parent-div-close

.. raw:: html

    </div>


.. toctree::
   :hidden:
   :includehidden:


   /gallery/audio/index.rst
   /gallery/image/index.rst
   /gallery/text/index.rst


.. only:: html

  .. container:: sphx-glr-footer sphx-glr-footer-gallery

    .. container:: sphx-glr-download sphx-glr-download-python

      :download:`Download all examples in Python source code: gallery_python.zip </gallery/gallery_python.zip>`

    .. container:: sphx-glr-download sphx-glr-download-jupyter

      :download:`Download all examples in Jupyter notebooks: gallery_jupyter.zip </gallery/gallery_jupyter.zip>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_