ERMLPInteraction¶
-
class
ERMLPInteraction(embedding_dim, hidden_dim)[source]¶ Bases:
pykeen.nn.modules.FunctionalInteraction[torch.FloatTensor,torch.FloatTensor,torch.FloatTensor]A stateful module for the ER-MLP interaction.
\[f(h, r, t) = W_2 ReLU(W_1 cat(h, r, t) + b_1) + b_2\]Initialize the interaction function.
- Parameters
Methods Summary
func(r, t, hidden, activation, final)Evaluate the ER-MLP interaction function.
Reset parameters the interaction function may have.
Methods Documentation
-
func(r, t, hidden, activation, final)¶ Evaluate the ER-MLP interaction function.
- Parameters
h (
FloatTensor) – shape: (batch_size, num_heads, 1, 1, dim) The head representations.r (
FloatTensor) – shape: (batch_size, 1, num_relations, 1, dim) The relation representations.t (
FloatTensor) – shape: (batch_size, 1, 1, num_tails, dim) The tail representations.hidden (
Linear) – The first linear layer.activation (
Module) – The activation function of the hidden layer.final (
Linear) – The second linear layer.
- Return type
FloatTensor- Returns
shape: (batch_size, num_heads, num_relations, num_tails) The scores.