DistMultLiteralGated

class DistMultLiteralGated(triples_factory: TriplesNumericLiteralsFactory, embedding_dim: int = 50, input_dropout: float = 0.0, **kwargs)[source]

Bases: LiteralModel

An implementation of the LiteralE model with thhe Gated DistMult interaction from [kristiadi2018].

This model is different from pykeen.models.DistMultLiteral because it uses a gate (like found in LSTMs) instead of a LinearDropout module.

This gate implements the full \(g\) function described in the LiteralE paper (see equation 4).

Initialize the model.

Parameters:

  • triples_factory (TriplesNumericLiteralsFactory) – the (training) triples factory

  • embedding_dim (int) – the embedding dimension

  • input_dropout (float) – the input dropout, cf. DistMultCombination.__init__()

  • kwargs – additional keyword-based parameters passed to LiteralModel.__init__()

Attributes Summary

hpo_default

The default strategy for optimizing the model's hyper-parameters

loss_default_kwargs

The default parameters for the default loss function class

Attributes Documentation

hpo_default: ClassVar[Mapping[str, Any]] = {'embedding_dim': {'high': 256, 'low': 16, 'q': 16, 'type': <class 'int'>}, 'input_dropout': {'high': 0.5, 'low': 0.0, 'q': 0.1, 'type': <class 'float'>}}

The default strategy for optimizing the model’s hyper-parameters

loss_default_kwargs: ClassVar[Mapping[str, Any]] = {'margin': 0.0}

The default parameters for the default loss function class