BestCheckpointSchedule

class BestCheckpointSchedule(result_tracker: ResultTracker, metric_selection: MetricSelection)[source]

Create a checkpoint whenever a metric improves.

Attributes Summary

`metric_selection`	the metric selection
`result_tracker`	the result tracker which receives updates on metrics since the same tracker instance needs to receive results from the training loop, we do require a pre-instantiated one rather than offering to provide hints, too

Methods Summary

Decide whether to create a checkpoint at the specified epoch.

Attributes Documentation

Parameters:

metric_selection: MetricSelection = <dataclasses._MISSING_TYPE object>: the metric selection

result_tracker: ResultTracker = <dataclasses._MISSING_TYPE object>: the result tracker which receives updates on metrics since the same tracker instance needs to receive results from the training loop, we do require a pre-instantiated one rather than offering to provide hints, too

Methods Documentation

__call__(step: int) → bool[source]

Decide whether to create a checkpoint at the specified epoch.