DVCLive allows you to easily add experiment tracking capabilities to your PyTorch Lightning projects.
PyTorch Lightning is an open-source framework for training PyTorch networks.
To start using DVCLive you just need to add a few lines to your training code in any PyTorch Lightning project.
+from dvclive.lightning import DvcLiveLogger . . . dvclive_logger = DvcLiveLogger() trainer = Trainer( + logger=dvclive_logger, ) trainer.fit(model)
This will generate the metrics logs and summaries 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.
run_name - (
None by default) - Name of the run, used in PyTorch Lightning
to get version.
prefix - (
None by default) - string that adds to each metric name.
experiment - (
None by default) -
Live object to be used instead of
initializing a new one.
**kwargs - Any additional arguments will be passed to
from dvclive.lightning import DvcLiveLogger dvclive_logger = DvcLiveLogger( path='my_logs_path', summary=False ) trainer = Trainer( logger=dvclive_logger, ) trainer.fit(model)
📖 By default, PyTorch Lightning creates a directory to store checkpoints using the logger's name (
DvcLiveLogger). You can change the checkpoint path or disable checkpointing at all as described in the PyTorch Lightning documentation