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
Fast-forward the training loop to record throughput for specific epochs and/or steps. |
|
Callback to log the L2 norm of all the parameters in the model, and optionally, the L2 norm of the parameters by each layer. |
|
Callback to log the learning rate. |
|
Callback to monitor the training throughput. |
|
Profile the execution using |