XLAProfiler

class lightning.pytorch.profilers.XLAProfiler(port=9012)[source]

Bases: Profiler

XLA Profiler will help you debug and optimize training workload performance for your models using Cloud TPU performance tools.

Parameters:

port (int) – the port to start the profiler server on. An exception is raised if the provided port is invalid or busy.

start(action_name)[source]

Defines how to start recording an action.

Return type:

None

stop(action_name)[source]

Defines how to record the duration once an action is complete.

Return type:

None