Stoppers¶
Early stoppers.
The following code will create a scenario in which training will stop
(quite) early when training pykeen.models.TransE
on the
pykeen.datasets.Nations
dataset.
>>> from pykeen.pipeline import pipeline
>>> pipeline_result = pipeline(
... dataset='nations',
... model='transe',
... model_kwargs=dict(embedding_dim=20, scoring_fct_norm=1),
... optimizer='SGD',
... optimizer_kwargs=dict(lr=0.01),
... loss='marginranking',
... loss_kwargs=dict(margin=1),
... training_loop='slcwa',
... training_kwargs=dict(num_epochs=100, batch_size=128),
... negative_sampler='basic',
... negative_sampler_kwargs=dict(num_negs_per_pos=1),
... evaluator_kwargs=dict(filtered=True),
... evaluation_kwargs=dict(batch_size=128),
... stopper='early',
... stopper_kwargs=dict(frequency=5, patience=2, delta=0.002),
... )
- class EarlyStopper(model, evaluator, evaluation_triples_factory, evaluation_batch_size=None, evaluation_slice_size=None, frequency=10, patience=2, metric='hits_at_k', relative_delta=0.01, best_metric=None, best_epoch=None, results=<factory>, larger_is_better=True, result_tracker=None, result_callbacks=<factory>, continue_callbacks=<factory>, stopped_callbacks=<factory>, stopped=False)[source]¶
A harness for early stopping.
- continue_callbacks: List[Callable[[pykeen.stoppers.stopper.Stopper, Union[int, float], int], None]]¶
Callbacks when training gets continued
- evaluation_triples_factory: Optional[pykeen.triples.triples_factory.TriplesFactory]¶
The triples to use for evaluation
- evaluator: pykeen.evaluation.evaluator.Evaluator¶
The evaluator
- model: pykeen.models.base.Model¶
The model
- property number_results: int¶
Count the number of results stored in the early stopper.
- Return type
- patience: int = 2¶
The number of iterations (one iteration can correspond to various epochs) with no improvement after which training will be stopped.
- relative_delta: float = 0.01¶
The minimum relative improvement necessary to consider it an improved result
- result_callbacks: List[Callable[[pykeen.stoppers.stopper.Stopper, Union[int, float], int], None]]¶
Callbacks when after results are calculated
- result_tracker: Optional[pykeen.trackers.ResultTracker] = None¶
The result tracker
- should_evaluate(epoch)[source]¶
Decide if evaluation should be done based on the current epoch and the internal frequency.
- Return type