class TransRInteraction(p, power_norm=True)[source]

Bases: pykeen.nn.modules.TranslationalInteraction[torch.FloatTensor, Tuple[torch.FloatTensor, torch.FloatTensor], torch.FloatTensor]

A stateful module for the TransR interaction function.

Initialize the translational interaction function.

  • p (int) – The norm used with torch.norm(). Typically is 1 or 2.

  • power_norm (bool) – Whether to use the p-th power of the L_p norm. It has the advantage of being differentiable around 0, and numerically more stable.

Attributes Summary


The symbolic shapes for relation representations

Methods Summary

func(r, t, m_r, p[, power_norm])

Evaluate the TransR interaction function.

Attributes Documentation

relation_shape: Sequence[str] = ('e', 'de')

The symbolic shapes for relation representations

Methods Documentation

func(r, t, m_r, p, power_norm=True)

Evaluate the TransR interaction function.

  • h (FloatTensor) – shape: (batch_size, num_heads, 1, 1, d_e) Head embeddings.

  • r (FloatTensor) – shape: (batch_size, 1, num_relations, 1, d_r) Relation embeddings.

  • m_r (FloatTensor) – shape: (batch_size, 1, num_relations, 1, d_e, d_r) The relation specific linear transformations.

  • t (FloatTensor) – shape: (batch_size, 1, 1, num_tails, d_e) Tail embeddings.

  • p (int) – The parameter p for selecting the norm.

  • power_norm (bool) – Whether to return the powered norm instead.

Return type



shape: (batch_size, num_heads, num_relations, num_tails) The scores.