Shortcuts

wandb

Classes

WandbLogger

Log using Weights and Biases.

Weights and Biases Logger

class pytorch_lightning.loggers.wandb.WandbLogger(name=None, save_dir=None, offline=False, id=None, anonymous=None, version=None, project=None, log_model=False, experiment=None, prefix='', **kwargs)[source]

Bases: pytorch_lightning.loggers.base.LightningLoggerBase

Log using Weights and Biases.

Installation and set-up

Install with pip:

pip install wandb

Create a WandbLogger instance:

from pytorch_lightning.loggers import WandbLogger

wandb_logger = WandbLogger(project="MNIST")

Pass the logger instance to the Trainer:

trainer = Trainer(logger=wandb_logger)

A new W&B run will be created when training starts if you have not created one manually before with wandb.init().

Log metrics

Log from LightningModule:

class LitModule(LightningModule):
    def training_step(self, batch, batch_idx):
        self.log("train/loss", loss)

Use directly wandb module:

wandb.log({"train/loss": loss})

Log hyper-parameters

Save LightningModule parameters:

class LitModule(LightningModule):
    def __init__(self, *args, **kwarg):
        self.save_hyperparameters()

Add other config parameters:

# add one parameter
wandb_logger.experiment.config["key"] = value

# add multiple parameters
wandb_logger.experiment.config.update({key1: val1, key2: val2})

# use directly wandb module
wandb.config["key"] = value
wandb.config.update()

Log gradients, parameters and model topology

Call the watch method for automatically tracking gradients:

# log gradients and model topology
wandb_logger.watch(model)

# log gradients, parameter histogram and model topology
wandb_logger.watch(model, log="all")

# change log frequency of gradients and parameters (100 steps by default)
wandb_logger.watch(model, log_freq=500)

# do not log graph (in case of errors)
wandb_logger.watch(model, log_graph=False)

The watch method adds hooks to the model which can be removed at the end of training:

wandb_logger.unwatch(model)

Log model checkpoints

Log model checkpoints at the end of training:

wandb_logger = WandbLogger(log_model=True)

Log model checkpoints as they get created during training:

wandb_logger = WandbLogger(log_model="all")

Custom checkpointing can be set up through ModelCheckpoint:

# log model only if `val_accuracy` increases
wandb_logger = WandbLogger(log_model="all")
checkpoint_callback = ModelCheckpoint(monitor="val_accuracy", mode="max")
trainer = Trainer(logger=wandb_logger, callbacks=[checkpoint_callback])

latest and best aliases are automatically set to easily retrieve a model checkpoint:

# reference can be retrieved in artifacts panel
# "VERSION" can be a version (ex: "v2") or an alias ("latest or "best")
checkpoint_reference = "USER/PROJECT/MODEL-RUN_ID:VERSION"

# download checkpoint locally (if not already cached)
run = wandb.init(project="MNIST")
artifact = run.use_artifact(checkpoint_reference, type="model")
artifact_dir = artifact.download()

# load checkpoint
model = LitModule.load_from_checkpoint(Path(artifact_dir) / "model.ckpt")

Log media

Log text with:

# using columns and data
columns = ["input", "label", "prediction"]
data = [["cheese", "english", "english"], ["fromage", "french", "spanish"]]
wandb_logger.log_text(key="samples", columns=columns, data=data)

# using a pandas DataFrame
wandb_logger.log_text(key="samples", dataframe=my_dataframe)

Log images with:

# using tensors, numpy arrays or PIL images
wandb_logger.log_image(key="samples", images=[img1, img2])

# adding captions
wandb_logger.log_image(key="samples", images=[img1, img2], caption=["tree", "person"])

# using file path
wandb_logger.log_image(key="samples", images=["img_1.jpg", "img_2.jpg"])

More arguments can be passed for logging segmentation masks and bounding boxes. Refer to Image Overlays documentation.

Log Tables

W&B Tables can be used to log, query and analyze tabular data.

They support any type of media (text, image, video, audio, molecule, html, etc) and are great for storing, understanding and sharing any form of data, from datasets to model predictions.

columns = ["caption", "image", "sound"]
data = [["cheese", wandb.Image(img_1), wandb.Audio(snd_1)], ["wine", wandb.Image(img_2), wandb.Audio(snd_2)]]
wandb_logger.log_table(key="samples", columns=columns, data=data)

See also

Parameters
  • name (Optional[str]) – Display name for the run.

  • save_dir (Optional[str]) – Path where data is saved (wandb dir by default).

  • offline (Optional[bool]) – Run offline (data can be streamed later to wandb servers).

  • id (Optional[str]) – Sets the version, mainly used to resume a previous run.

  • version (Optional[str]) – Same as id.

  • anonymous (Optional[bool]) – Enables or explicitly disables anonymous logging.

  • project (Optional[str]) – The name of the project to which this run will belong.

  • log_model (Union[str, bool]) –

    Log checkpoints created by ModelCheckpoint as W&B artifacts. latest and best aliases are automatically set.

    • if log_model == 'all', checkpoints are logged during training.

    • if log_model == True, checkpoints are logged at the end of training, except when save_top_k == -1 which also logs every checkpoint during training.

    • if log_model == False (default), no checkpoint is logged.

  • prefix (Optional[str]) – A string to put at the beginning of metric keys.

  • experiment – WandB experiment object. Automatically set when creating a run.

  • **kwargs – Arguments passed to wandb.init() like entity, group, tags, etc.

Raises
  • ModuleNotFoundError – If required WandB package is not installed on the device.

  • MisconfigurationException – If both log_model and offline is set to True.

after_save_checkpoint(checkpoint_callback)[source]

Called after model checkpoint callback saves a new checkpoint.

Parameters

checkpoint_callback – the model checkpoint callback instance

finalize(status)[source]

Do any processing that is necessary to finalize an experiment.

Parameters

status (str) – Status that the experiment finished with (e.g. success, failed, aborted)

Return type

None

log_hyperparams(params)[source]

Record hyperparameters.

Parameters
  • params (Union[Dict[str, Any], Namespace]) – Namespace containing the hyperparameters

  • args – Optional positional arguments, depends on the specific logger being used

  • kwargs – Optional keyword arguments, depends on the specific logger being used

Return type

None

log_image(key, images, step=None, **kwargs)[source]

Log images (tensors, numpy arrays, PIL Images or file paths).

Optional kwargs are lists passed to each image (ex: caption, masks, boxes).

Return type

None

log_metrics(metrics, step=None)[source]

Records metrics. This method logs metrics as as soon as it received them. If you want to aggregate metrics for one specific step, use the agg_and_log_metrics() method.

Parameters
  • metrics (Dict[str, float]) – Dictionary with metric names as keys and measured quantities as values

  • step (Optional[int]) – Step number at which the metrics should be recorded

Return type

None

log_table(key, columns=None, data=None, dataframe=None, step=None)[source]

Log a Table containing any object type (text, image, audio, video, molecule, html, etc).

Can be defined either with columns and data or with dataframe.

Return type

None

log_text(key, columns=None, data=None, dataframe=None, step=None)[source]

Log text as a Table.

Can be defined either with columns and data or with dataframe.

Return type

None

property experiment: wandb.wandb_run.Run

Actual wandb object. To use wandb features in your LightningModule do the following.

Example:

.. code-block:: python

self.logger.experiment.some_wandb_function()

Return type

Run

property name: Optional[str]

Gets the name of the experiment.

Return type

Optional[str]

Returns

The name of the experiment if the experiment exists else the name given to the constructor.

property save_dir: Optional[str]

Gets the save directory.

Return type

Optional[str]

Returns

The path to the save directory.

property version: Optional[str]

Gets the id of the experiment.

Return type

Optional[str]

Returns

The id of the experiment if the experiment exists else the id given to the constructor.