PyTorch Inference Monitoring and Profiling

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

Add the following code around inference. See API reference for full documentation.

with graphsignal.start_trace(endpoint='predict', profiler='pytorch'):
    # function call or code segment

Examples

The PyTorch MNIST example illustrates where and how to add the start_trace method.

Model serving

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

Useful links

• Performance Tuning Guide
• Accelerate PyTorch Model Inferencing