PyTorch Lightning Inference Monitoring and Profiling

See the Quick Start guide on how to install and configure Graphsignal.

Use a Trainer callback for validation, test and prediction phases:

from graphsignal.callbacks.pytorch_lightning import GraphsignalCallback

trainer = Trainer(..., callbacks=GraphsignalCallback(endpoint='predict')])
trainer.predict() # or trainer.validate() or trainer.test()

Note: avoid using other PyTorch Lightning profilers simultaneously.

Examples

The PyTorch Lightning MNIST example illustrates how to add the callback.

Model serving

Graphsignal provides a built-in support for server applications. See Model Serving guide for more information.

Useful links

• Inference in Production