LCWAEvaluationLoop

class LCWAEvaluationLoop(triples_factory: CoreTriplesFactory, evaluator: str | Evaluator | type[Evaluator] | None = None, evaluator_kwargs: Mapping[str, Any] | None = None, targets: Collection[Literal['head', 'relation', 'tail']] = ('head', 'tail'), mode: Literal['training', 'validation', 'testing'] | None = None, additional_filter_triples: Tensor | CoreTriplesFactory | Sequence[Tensor | CoreTriplesFactory] | None = None, **kwargs)[source]

Bases: EvaluationLoop[Mapping[Literal[‘head’, ‘relation’, ‘tail’], Tensor]]

Evaluation loop using 1:n scoring.

For brevity, we only describe evaluation for tail prediction. Let \((h, r, t) \in \mathcal{T}_{eval}\) denote an evaluation triple. Then, we calculate scores for all triples \((h, r, t')\) with \(t' \in \mathcal{E}\), i.e., for replacing the true tail \(t\) by all entities.

Initialize the evaluation loop.

Parameters:
  • triples_factory (CoreTriplesFactory) – the evaluation triples factory

  • evaluator (str | Evaluator | type[Evaluator] | None) – the evaluator, or a hint thereof

  • evaluator_kwargs (Mapping[str, Any] | None) – additional keyword-based parameters for instantiating the evaluator

  • targets (Collection[Literal['head', 'relation', 'tail']]) – the prediction targets.

  • mode (Literal['training', 'validation', 'testing'] | None) – the inductive mode, or None for transductive evaluation

  • additional_filter_triples (Tensor | CoreTriplesFactory | Sequence[Tensor | CoreTriplesFactory] | None) – additional filter triples to use for creating the filter

  • kwargs – additional keyword-based parameters passed to EvaluationLoop.__init__(). Should not contain the keys dataset or evaluator.

Methods Summary

get_collator()

Get the collator to use for the data loader.

process_batch(batch)

Process a single batch.

Methods Documentation

get_collator()[source]

Get the collator to use for the data loader.

process_batch(batch: Mapping[Literal['head', 'relation', 'tail'], Tensor]) None[source]

Process a single batch.

Parameters:

batch (Mapping[Literal['head', 'relation', 'tail'], ~torch.Tensor]) – one batch of evaluation samples from the dataset.

Return type:

None