NGCF¶

Reference:: Xiang Wang et al. “Neural Graph Collaborative Filtering.” in SIGIR 2019.
Reference code:: https://github.com/xiangwang1223/neural_graph_collaborative_filtering

class recbole.model.general_recommender.ngcf.NGCF(config, dataset)[source]¶

Bases: recbole.model.abstract_recommender.GeneralRecommender

NGCF is a model that incorporate GNN for recommendation. We implement the model following the original author with a pairwise training mode.

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

forward()[source]¶

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)[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

get_ego_embeddings()[source]¶

Get the embedding of users and items and combine to an embedding matrix.

Returns: Tensor of the embedding matrix. Shape of (n_items+n_users, embedding_dim)

get_eye_mat()[source]¶

Construct the identity matrix with the size of n_items+n_users.

Returns: Sparse tensor of the identity matrix. Shape of (n_items+n_users, n_items+n_users)

get_norm_adj_mat()[source]¶

Get the normalized interaction matrix of users and items.

Construct the square matrix from the training data and normalize it using the laplace matrix.

\[A_{hat} = D^{-0.5} \times A \times D^{-0.5}\]

Returns: Sparse tensor of the normalized interaction matrix.

input_type = 2¶

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

recbole.model.general_recommender.ngcf.sparse_dropout(x, rate, noise_shape)[source]¶

This is a function that execute Dropout on Pytorch sparse tensor.

A random dropout will be applied to the input sparse tensor.

Note

input tensor SHOULD be a sparse float tensor. we suggest to use ‘._nnz()’ as the shape of sparse tensor for an easy calling.

Parameters

x (torch.sparse.FloatTensor) – The input sparse tensor.
rate (float) – Dropout rate which should in [0,1].
noise_shape (tuple) – Shape of the input sparse tensor. suggest ‘._nnz()’

Returns

The result sparse tensor after dropout.

Return type

torch.sparse.FloatTensor