Automatically log PyTorchLightning metrics to Neptune

  • Log PyTorch Lightning metrics to Neptune
  • Use PyTorch Lightning lightweight API and track deep learning experiments in Neptune
  • Gain insights from comparing experiments interactively
Experiments tracking
Experiments sharing in the team
Experiments comparison
Collaboration
Notebooks tracking and sharing
Notebooks comparison
Team management
Open source integrations
Hardware monitoring for experiments
Interactive experiments dashboard

Automatically track deep learning training with PyTorch Lightning in Neptune

Track model training progress to Neptune while using lightweight PyTorch Lightning API.

Neptune integration with PyTorch Lightning is part of the PyTorch Lightning library. To track deep learning experiments simply use NeptuneLogger from the logging module. Track: metrics, images, model weights, parameters and more. Organize your experiments with tags.

 

from pytorch_lightning import Trainer
from pytorch_lightning.logging.neptune import NeptuneLogger

neptune_logger = NeptuneLogger(
    api_key="ANONYMOUS",
    project_name="shared/pytorch-lightning-integration",
    params={"max_epochs": 10,
            "batch_size": 32},
)
model = CoolSystem()

# Use neptune_logger here
trainer = Trainer(max_epochs=10, logger=neptune_logger)
trainer.fit(model)

Support for deep learning libraries

  • Start instantly with out-of-the-box integration.
  • Track rich data (metrics, text, images, files and more).
  • Save model weights.

Shareable visualizations

  • All charts visible for team members.
  • Compare across multiple experiments and gain insight.
  • Download charts from UI.

Backed up experiments history

  • All visualizations are stored and secured.
  • Keep all history – review when needed.
  • Secured Intellectual Property (IP).

Register and try it out!