composer.callbacks

Note

To write a custom callback, see composer.Callback.

Composer contains built-in callbacks, which can be added via the --callbacks CLI flag:

python examples/run_mosaic_trainer.py -f my_model.yaml --callbacks lr_monitor grad_monitor

Callbacks

`Benchmarker`	Fast-forward the training loop to record throughput for specific epochs and/or steps.
`GradMonitor`	Callback to log the L2 norm of all the parameters in the model, and optionally, the L2 norm of the parameters by each layer.
`LRMonitor`	Callback to log the learning rate.
`SpeedMonitor`	Callback to monitor the training throughput.
`TorchProfiler`	Profile the execution using `torch.profiler.profile`.