hopwise.model.knowledge_graph_embedding_recommender.complex¶

Classes¶

Module Contents¶

class hopwise.model.knowledge_graph_embedding_recommender.complex.ComplEx(config, dataset)¶

Bases: hopwise.model.abstract_recommender.KnowledgeRecommender

ComplEx extends DistMult by introducing complex-valued embeddings.

Note

In this version, we sample recommender data and knowledge data separately, and put them together for training.

input_type¶

embedding_size¶

device¶

ui_relation¶

user_re_embedding¶

user_im_embedding¶

entity_re_embedding¶

entity_im_embedding¶

relation_re_embedding¶

relation_im_embedding¶

loss¶

forward(head_re_e, head_im_e, rec_r_re_e, rec_r_im_e, tail_re_e, tail_im_e)¶

triple_dot(x, y, z)¶

_get_rec_embeddings(user, positive_items, negative_items)¶

_get_kg_embeddings(head, relation, positive_tails, negative_tails)¶

calculate_loss(interaction)¶

Calculate the training loss for a batch data.

Parameters:

interaction (Interaction) – Interaction class of the batch.

Returns:

Training loss, shape: []

Return type:

torch.Tensor

predict(interaction)¶

Predict the scores between users and items.

Parameters:

interaction (Interaction) – Interaction class of the batch.

Returns:

Predicted scores for given users and items, shape: [batch_size]

Return type:

torch.Tensor

predict_kg(interaction)¶

full_sort_predict(interaction)¶

Full sort prediction function. Given users, calculate the scores between users and all candidate items.

Parameters:

interaction (Interaction) – Interaction class of the batch.

Returns:

Predicted scores for given users and all candidate items, shape: [n_batch_users * n_candidate_items]

Return type:

torch.Tensor

full_sort_predict_kg(interaction)¶

Full sort prediction KG function. Given heads, calculate the scores between heads and all candidate tails.

Parameters:

interaction (Interaction) – Interaction class of the batch.

Returns:

Predicted scores for given heads and all candidate tails, shape: [n_batch_heads * n_candidate_tails]

Return type:

torch.Tensor