composer.callbacks#
Callbacks that run at each training loop Event
.
Each callback inherits from the Callback
base class. See detailed description and
examples for writing your own callbacks at the Callback
base class.
Classes
Logs stats of activation inputs and outputs. |
|
Callback to save checkpoints. |
|
Track a metric and halt training if it does not improve within a given interval. |
|
Callback to export model for inference. |
|
Free train metrics on AFTER_LOSS to reduce peak memory usage if not using train metrics. |
|
Periodically log generations from a set of prompts. |
|
Logs image inputs and optionally outputs. |
|
Logs the learning rate. |
|
Create compliant results file for MLPerf Training benchmark. |
|
Logs the memory usage of the model. |
|
Logs the memory snapshot of the model. |
|
Catches NaNs in the loss and raises an error if one is found. |
|
Generate visualizations of the state of allocated memory during an OutOfMemory exception. |
|
Computes and logs the L2 norm of gradients as well as any optimizer-specific metrics implemented in the optimizer's report_per_parameter_metrics method. |
|
Estimates total training time. |
|
Logs the training throughput and utilization. |
|
Logs GPU/CPU metrics. |
|
Halt training when a metric value reaches a certain threshold. |