hopwise.model.knowledge_aware_recommender.mkr¶

Reference:: Hongwei Wang et al. “Multi-Task Feature Learning for Knowledge Graph Enhanced Recommendation.” in WWW 2019.
Reference code:: https://github.com/hsientzucheng/MKR.PyTorch

Classes¶

`MKR`	MKR is a Multi-task feature learning approach for Knowledge graph enhanced Recommendation. It is a deep
`CrossCompressUnit`	This is Cross&Compress Unit for MKR model to model feature interactions between items and entities.

Module Contents¶

class hopwise.model.knowledge_aware_recommender.mkr.MKR(config, dataset)[source]¶

Bases: hopwise.model.abstract_recommender.KnowledgeRecommender

MKR is a Multi-task feature learning approach for Knowledge graph enhanced Recommendation. It is a deep end-to-end framework that utilizes knowledge graph embedding task to assist recommendation task. The two tasks are associated by cross&compress units, which automatically share latent features and learn high-order interactions between items in recommender systems and entities in the knowledge graph.

input_type¶

LABEL¶

embedding_size¶

kg_embedding_size¶

L¶

H¶

reg_weight¶

use_inner_product¶

dropout_prob¶

user_embeddings_lookup¶

item_embeddings_lookup¶

entity_embeddings_lookup¶

relation_embeddings_lookup¶

user_mlp¶

tail_mlp¶

cc_unit¶

kge_mlp¶

kge_pred_mlp¶

sigmoid_BCE¶

forward(user_indices=None, item_indices=None, head_indices=None, relation_indices=None, tail_indices=None)[source]¶

_l2_loss(inputs)[source]¶

calculate_rs_loss(interaction)[source]¶: Calculate the training loss for a batch data of RS.

calculate_kg_loss(interaction)[source]¶: Calculate the training loss for a batch data of KG.

predict(interaction)[source]¶

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

class hopwise.model.knowledge_aware_recommender.mkr.CrossCompressUnit(dim)[source]¶

Bases: torch.nn.Module

This is Cross&Compress Unit for MKR model to model feature interactions between items and entities.

dim¶

fc_vv¶

fc_ev¶

fc_ve¶

fc_ee¶

forward(inputs)[source]¶