StructuredEmbedding

class StructuredEmbedding(*, embedding_dim=50, scoring_fct_norm=1, entity_initializer=<function xavier_uniform_>, entity_constrainer=<function normalize>, entity_constrainer_kwargs=None, **kwargs)

Bases: pykeen.models.base.EntityEmbeddingModel

An implementation of the Structured Embedding (SE) published by [bordes2011].

SE applies role- and relation-specific projection matrices \(\textbf{M}_{r}^{h}, \textbf{M}_{r}^{t} \in \mathbb{R}^{d \times d}\) to the head and tail entities’ embeddings before computing their differences. Then, the \(l_p\) norm is applied and the result is negated such that smaller differences are considered better.

\[f(h, r, t) = - \|\textbf{M}_{r}^{h} \textbf{e}_h - \textbf{M}_{r}^{t} \textbf{e}_t\|_p\]

By employing different projections for the embeddings of the head and tail entities, SE explicitly differentiates the role of an entity as either the subject or object.

Initialize SE.

Parameters

• embedding_dim (int) – The entity embedding dimension \(d\). Is usually \(d \in [50, 300]\).

• scoring_fct_norm (int) – The \(l_p\) norm. Usually 1 for SE.

• entity_initializer (Union[str, Callable[[FloatTensor], FloatTensor], None]) – Entity initializer function. Defaults to pykeen.nn.init.xavier_uniform_()

• entity_constrainer (Union[str, Callable[[FloatTensor], FloatTensor], None]) – Entity constrainer function. Defaults to torch.nn.functional.normalize()

• entity_constrainer_kwargs (Optional[Mapping[str, Any]]) – Keyword arguments to be used when calling the entity constrainer

• kwargs – Remaining keyword arguments to forward to pykeen.models.EntityEmbeddingModel

Attributes Summary

hpo_default

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

Methods Summary

score_h(rt_batch[, slice_size])	Forward pass using left side (head) prediction.
score_hrt(hrt_batch)	Forward pass.
score_t(hr_batch[, slice_size])	Forward pass using right side (tail) prediction.

Attributes Documentation

hpo_default: ClassVar[Mapping[str, Any]] = {'embedding_dim': {'high': 256, 'low': 16, 'q': 16, 'type': <class 'int'>}, 'scoring_fct_norm': {'high': 2, 'low': 1, 'type': <class 'int'>}}

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

Methods Documentation

score_h(rt_batch, slice_size=None)

Forward pass using left side (head) prediction.

This method calculates the score for all possible heads for each (relation, tail) pair.

Parameters

rt_batch (LongTensor) – shape: (batch_size, 2), dtype: long The indices of (relation, tail) pairs.

Return type

FloatTensor

Returns

shape: (batch_size, num_entities), dtype: float For each r-t pair, the scores for all possible heads.

score_hrt(hrt_batch)

Forward pass.

This method takes head, relation and tail of each triple and calculates the corresponding score.

Parameters

hrt_batch (LongTensor) – shape: (batch_size, 3), dtype: long The indices of (head, relation, tail) triples.

Raises

NotImplementedError – If the method was not implemented for this class.

Return type

FloatTensor

Returns

shape: (batch_size, 1), dtype: float The score for each triple.

score_t(hr_batch, slice_size=None)

Forward pass using right side (tail) prediction.

This method calculates the score for all possible tails for each (head, relation) pair.

Parameters

hr_batch (LongTensor) – shape: (batch_size, 2), dtype: long The indices of (head, relation) pairs.

Return type

FloatTensor

Returns

shape: (batch_size, num_entities), dtype: float For each h-r pair, the scores for all possible tails.