GCSAN

Reference:

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

class recbole.model.sequential_recommender.gcsan.GCSAN(config, dataset)

Bases: recbole.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.

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

forward(item_seq, item_seq_len)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

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

get_attention_mask(item_seq)

Generate left-to-right uni-directional attention mask for multi-head attention.

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

training: bool

class recbole.model.sequential_recommender.gcsan.GNN(embedding_size, step=1)

Bases: torch.nn.modules.module.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.

GNNCell(A, hidden)

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)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

training: bool