lightning.rst

import torch from torch.nn import Module from pytorch_lightning.core.lightning import LightningModule from torchmetrics import Metric

TorchMetrics in PyTorch Lightning

TorchMetrics was originally created as part of PyTorch Lightning, a powerful deep learning research framework designed for scaling models without boilerplate.

Note

TorchMetrics always offers compatibility with the last 2 major PyTorch Lightning versions, but we recommend to always keep both frameworks up-to-date for the best experience.

While TorchMetrics was built to be used with native PyTorch, using TorchMetrics with Lightning offers additional benefits:

• Modular metrics are automatically placed on the correct device when properly defined inside a LightningModule. This means that your data will always be placed on the same device as your metrics. No need to call .to(device) anymore!
• Native support for logging metrics in Lightning using self.log inside your LightningModule.
• The .reset() method of the metric will automatically be called at the end of an epoch.

The example below shows how to use a metric in your LightningModule:

python

class MyModel(LightningModule):

def __init__(self, num_classes):

... self.accuracy = torchmetrics.Accuracy(task="multiclass", num_classes=num_classes)

def training_step(self, batch, batch_idx):

x, y = batch preds = self(x) ... # log step metric self.accuracy(preds, y) self.log('train_acc_step', self.accuracy) ...

def training_epoch_end(self, outs):

# log epoch metric self.log('train_acc_epoch', self.accuracy)

Metric logging in Lightning happens through the self.log or self.log_dict method. Both methods only support the logging of scalar-tensors. While the vast majority of metrics in torchmetrics returns a scalar tensor, some metrics such as ~torchmetrics.ConfusionMatrix, ~torchmetrics.ROC, ~torchmetrics.MeanAveragePrecision, ~torchmetrics.ROUGEScore return outputs that are non-scalar tensors (often dicts or list of tensors) and should therefore be dealt with separately. For info about the return type and shape please look at the documentation for the compute method for each metric you want to log.

Logging TorchMetrics

Logging metrics can be done in two ways: either logging the metric object directly or the computed metric values. When ~torchmetrics.Metric objects, which return a scalar tensor are logged directly in Lightning using the LightningModule self.log method, Lightning will log the metric based on on_step and on_epoch flags present in self.log(...). If on_epoch is True, the logger automatically logs the end of epoch metric value by calling .compute().

Note

sync_dist, sync_dist_op, sync_dist_group, reduce_fx and tbptt_reduce_fx flags from self.log(...) don't affect the metric logging in any manner. The metric class contains its own distributed synchronization logic.

This however is only true for metrics that inherit the base class Metric, and thus the functional metric API provides no support for in-built distributed synchronization or reduction functions.

python

class MyModule(LightningModule):

def __init__(self, num_classes):

... self.train_acc = torchmetrics.Accuracy(task="multiclass", num_classes=num_classes) self.valid_acc = torchmetrics.Accuracy(task="multiclass", num_classes=num_classes)

def training_step(self, batch, batch_idx):

x, y = batch preds = self(x) ... self.train_acc(preds, y) self.log('train_acc', self.train_acc, on_step=True, on_epoch=False)

def validation_step(self, batch, batch_idx):

logits = self(x) ... self.valid_acc(logits, y) self.log('valid_acc', self.valid_acc, on_step=True, on_epoch=True)

As an alternative to logging the metric object and letting Lightning take care of when to reset the metric etc. you can also manually log the output of the metrics.

python

class MyModule(LightningModule):

def __init__(self):

... self.train_acc = torchmetrics.Accuracy(task="multiclass", num_classes=num_classes) self.valid_acc = torchmetrics.Accuracy(task="multiclass", num_classes=num_classes)

def training_step(self, batch, batch_idx):

x, y = batch preds = self(x) ... batch_value = self.train_acc(preds, y) self.log('train_acc_step', batch_value)

def training_epoch_end(self, outputs):

self.train_acc.reset()

def validation_step(self, batch, batch_idx):

logits = self(x) ... self.valid_acc.update(logits, y)

def validation_epoch_end(self, outputs):

self.log('valid_acc_epoch', self.valid_acc.compute()) self.valid_acc.reset()

Note that logging metrics this way will require you to manually reset the metrics at the end of the epoch yourself. In general, we recommend logging the metric object to make sure that metrics are correctly computed and reset. Additionally, we highly recommend that the two ways of logging are not mixed as it can lead to wrong results.

Note

When using any Modular metric, calling self.metric(...) or self.metric.forward(...) serves the dual purpose of calling self.metric.update() on its input and simultaneously returning the metric value over the provided input. So if you are logging a metric only on epoch-level (as in the example above), it is recommended to call self.metric.update() directly to avoid the extra computation.

python

class MyModule(LightningModule):

def __init__(self, num_classes):

... self.valid_acc = torchmetrics.Accuracy(task="multiclass", num_classes=num_classes)

def validation_step(self, batch, batch_idx):

logits = self(x) ... self.valid_acc.update(logits, y) self.log('valid_acc', self.valid_acc, on_step=True, on_epoch=True)

Common Pitfalls

The following contains a list of pitfalls to be aware of:

• If using metrics in data parallel mode (dp), the metric update/logging should be done in the <mode>_step_end method (where <mode> is either training, validation or test). This is because dp split the batches during the forward pass and metric states are destroyed after each forward pass, thus leading to wrong accumulation. In practice do the following:

python

class MyModule(LightningModule):

def training_step(self, batch, batch_idx):

data, target = batch preds = self(data) # ... return {'loss': loss, 'preds': preds, 'target': target}

def training_step_end(self, outputs):

# update and log self.metric(outputs['preds'], outputs['target']) self.log('metric', self.metric)

• Modular metrics contain internal states that should belong to only one DataLoader. In case you are using multiple DataLoaders, it is recommended to initialize a separate modular metric instances for each DataLoader and use them separately. The same holds for using seperate metrics for training, validation and testing.

python

class MyModule(LightningModule):

def __init__(self, num_classes):

... self.val_acc = nn.ModuleList([torchmetrics.Accuracy(task="multiclass", num_classes=num_classes) for _ in range(2)])

def val_dataloader(self):

return [DataLoader(...), DataLoader(...)]

def validation_step(self, batch, batch_idx, dataloader_idx):

x, y = batch preds = self(x) ... self.val_acc[dataloader_idx](preds, y) self.log('val_acc', self.val_acc[dataloader_idx])

• Mixing the two logging methods by calling self.log("val", self.metric) in {training}/{val}/{test}_step method and then calling self.log("val", self.metric.compute()) in the corresponding {training}/{val}/{test}_epoch_end method. Because the object is logged in the first case, Lightning will reset the metric before calling the second line leading to errors or nonsense results.
• Calling self.log("val", self.metric(preds, target)) with the intention of logging the metric object. Because self.metric(preds, target) corresponds to calling the forward method, this will return a tensor and not the metric object. Such logging will be wrong in this case. Instead it is important to seperate into seperate lines:

python

def training_step(self, batch, batch_idx):

x, y = batch preds = self(x) ... # log step metric self.accuracy(preds, y) # compute metrics self.log('train_acc_step', self.accuracy) # log metric object