hopwise.model.sequential_recommender.gcsan¶

Reference:: Chengfeng Xu et al. “Graph Contextualized Self-Attention Network for Session-based Recommendation.” in IJCAI 2019.

Classes¶

`GNN`	Graph neural networks are well-suited for session-based recommendation,
`GCSAN`	GCSAN captures rich local dependencies via graph neural network,

Module Contents¶

class hopwise.model.sequential_recommender.gcsan.GNN(embedding_size, step=1)[source]¶

Bases: torch.nn.Module

Graph neural networks are well-suited for session-based recommendation, because it can automatically extract features of session graphs with considerations of rich node connections.

step = 1¶

embedding_size¶

input_size¶

gate_size¶

w_ih¶

w_hh¶

b_ih¶

b_hh¶

linear_edge_in¶

linear_edge_out¶

_reset_parameters()[source]¶

GNNCell(A, hidden)[source]¶

Obtain latent vectors of nodes via gated graph neural network.

Parameters:

A (torch.FloatTensor) – The connection matrix,shape of [batch_size, max_session_len, 2 * max_session_len]
hidden (torch.FloatTensor) – The item node embedding matrix, shape of [batch_size, max_session_len, embedding_size]

Returns:

Latent vectors of nodes,shape of [batch_size, max_session_len, embedding_size]

Return type:

torch.FloatTensor

forward(A, hidden)[source]¶

class hopwise.model.sequential_recommender.gcsan.GCSAN(config, dataset)[source]¶

Bases: hopwise.model.abstract_recommender.SequentialRecommender

GCSAN captures rich local dependencies via graph neural network,: and learns long-range dependencies by applying the self-attention mechanism.

Note

In the original paper, the attention mechanism in the self-attention layer is a single head, for the reusability of the project code, we use a unified transformer component. According to the experimental results, we only applied regularization to embedding.

n_layers¶

n_heads¶

hidden_size¶

inner_size¶

hidden_dropout_prob¶

attn_dropout_prob¶

hidden_act¶

layer_norm_eps¶

step¶

device¶

weight¶

reg_weight¶

loss_type¶

initializer_range¶

item_embedding¶

gnn¶

self_attention¶

reg_loss¶

_init_weights(module)[source]¶: Initialize the weights

_get_slice(item_seq)[source]¶

forward(item_seq, item_seq_len)[source]¶

calculate_loss(interaction)[source]¶

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)[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

full_sort_predict(interaction)[source]¶

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