hopwise.model.knowledge_aware_recommender.kgin¶

Reference:: Xiang Wang et al. “Learning Intents behind Interactions with Knowledge Graph for Recommendation.” in WWW 2021.
Reference code:: https://github.com/huangtinglin/Knowledge_Graph_based_Intent_Network

Classes¶

`Aggregator`	Relational Path-aware Convolution Network
`GraphConv`	Graph Convolutional Network
`KGIN`	KGIN is a knowledge-aware recommendation model. It combines knowledge graph and the user-item interaction

Module Contents¶

class hopwise.model.knowledge_aware_recommender.kgin.Aggregator¶

Bases: torch.nn.Module

Relational Path-aware Convolution Network

forward(entity_emb, user_emb, latent_emb, relation_emb, edge_index, edge_type, interact_mat, disen_weight_att)¶

class hopwise.model.knowledge_aware_recommender.kgin.GraphConv(embedding_size, n_hops, n_users, n_factors, n_relations, edge_index, edge_type, interact_mat, ind, tmp, device, node_dropout_rate=0.5, mess_dropout_rate=0.1)¶

Bases: torch.nn.Module

Graph Convolutional Network

embedding_size¶

n_hops¶

n_relations¶

n_users¶

n_factors¶

edge_index¶

edge_type¶

interact_mat¶

node_dropout_rate = 0.5¶

mess_dropout_rate = 0.1¶

ind¶

temperature¶

device¶

relation_embedding¶

disen_weight_att¶

convs¶

node_dropout¶

mess_dropout¶

edge_sampling(edge_index, edge_type, rate=0.5)¶

forward(user_emb, entity_emb, latent_emb)¶: Node dropout

calculate_cor_loss(tensors)¶

class hopwise.model.knowledge_aware_recommender.kgin.KGIN(config, dataset)¶

Bases: hopwise.model.abstract_recommender.KnowledgeRecommender

KGIN is a knowledge-aware recommendation model. It combines knowledge graph and the user-item interaction graph to a new graph called collaborative knowledge graph (CKG). This model explores intents behind a user-item interaction by using auxiliary item knowledge.

input_type¶

embedding_size¶

n_factors¶

context_hops¶

node_dropout_rate¶

mess_dropout_rate¶

ind¶

sim_decay¶

reg_weight¶

temperature¶

interact_mat¶

kg_graph¶

n_nodes¶

user_embedding¶

entity_embedding¶

latent_embedding¶

gcn¶

mf_loss¶

reg_loss¶

restore_user_e = None¶

restore_entity_e = None¶

get_edges(graph)¶

forward()¶

calculate_loss(interaction)¶

Calculate the training loss for a batch data of KG.

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

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