composer.profiler#

Performance profiling tools.

The profiler gathers performance metrics during a training run that can be used to diagnose bottlenecks and facilitate model development.

The metrics gathered include:

  • Duration of each Event during training

  • Time taken by the data loader to return a batch

  • Host metrics such as CPU, system memory, disk and network utilization over time

  • Execution order, latency and attributes of PyTorch operators and GPU kernels (see torch.profiler)

See the Profiling Guide for additional information.

Functions

cyclic_schedule

Profiler schedule function for a cyclic profiling window.

Classes

JSONTraceHandler

Records trace events in Chrome JSON trace format.

Marker

Profiler Marker.

Profiler

Composer Profiler.

ProfilerAction

Action states for the Profiler that define whether or not events are being recorded to the trace file.

SystemProfiler

The SystemProfiler records system level metrics.

TorchProfiler

Profile the execution using the PyTorch Profiler.

TraceHandler

Base class for Composer Profiler trace handlers.