Skip to content

by iterative.ai

Catalyst

DVCLive allows you to add experiment tracking capabilities to your Catalyst projects.

Usage

To start using DVCLive, add a few lines to your training code in any Catalyst project.

Include the DvcLiveCallback int the callbacks list passed to your Runner:

+from dvclive.catalyst import DvcLiveCallback

. . .

runner.train(
    model=model,
    criterion=criterion,
    optimizer=optimizer,
    loaders=loaders,
-    num_epochs=2)
+    num_epochs=2,
+    callbacks=[DvcLiveCallback()])

This will generate the outputs as described in the Get Started.

Without requiring additional modifications to your training code, you can use DVCLive alongside DVC. See DVCLive with DVC for more info.

Parameters

model_file - (None by default) - The name of the file where the model will be saved at the end of each step.
**kwargs - Any additional arguments will be passed to Live.

Examples

Using model_file.

from dvclive.catalyst import DvcLiveCallback

runner.train(
    model=model,
    criterion=criterion,
    optimizer=optimizer,
    loaders=loaders,
    num_epochs=2,
    callbacks=[DvcLiveCallback(model_file="model.pth")])

Using **kwargs to customize Live.

from dvclive.catalyst import DvcLiveCallback

runner.train(
    model=model,
    criterion=criterion,
    optimizer=optimizer,
    loaders=loaders,
    num_epochs=2,
    callbacks=[
      DvcLiveCallback(model_file="model.pth", path="custom_path")])

🐛 Found an issue? Let us know! Or fix it:

❓ Have a question? Join our chat, we will help you: