pytorch_pfn_extras.training.triggers.BestValueTrigger¶
- class pytorch_pfn_extras.training.triggers.BestValueTrigger(key, compare, trigger=(1, 'epoch'))¶
Trigger invoked when specific value becomes best.
- Parameters
key (str) – Key of value.
compare (callable) – Compare function which takes current best value and new value and returns whether new value is better than current best.
trigger (TriggerLike) – Trigger that decides the comparison interval between current best value and new value. This must be a tuple in the form of
<int>, 'epoch'
or<int>, 'iteration'
which is passed toIntervalTrigger
.
- Return type
None
- __init__(key, compare, trigger=(1, 'epoch'))¶
- Parameters
key (str) –
compare (Callable[[float, float], bool]) –
trigger (TriggerLike) –
- Return type
None
Methods
__init__
(key, compare[, trigger])load_state_dict
(to_load)may_fire
(iteration, epoch_length)Flags if the trigger may fire at the current iteration
state_dict
()