RecVAE

Reference:

Shenbin, Ilya, et al. “RecVAE: A new variational autoencoder for Top-N recommendations with implicit feedback.” In WSDM 2020.

Reference code:

https://github.com/ilya-shenbin/RecVAE

class recbole.model.general_recommender.recvae.CompositePrior(hidden_dim, latent_dim, input_dim, mixture_weights)

Bases: torch.nn.modules.module.Module

forward(x, z)

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

class recbole.model.general_recommender.recvae.Encoder(hidden_dim, latent_dim, input_dim, eps=0.1)

Bases: torch.nn.modules.module.Module

forward(x, dropout_prob)

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

class recbole.model.general_recommender.recvae.RecVAE(config, dataset)

Bases: recbole.model.abstract_recommender.GeneralRecommender, recbole.model.abstract_recommender.AutoEncoderMixin

Collaborative Denoising Auto-Encoder (RecVAE) is a recommendation model for top-N recommendation with implicit feedback.

We implement the model following the original author

calculate_loss(interaction, encoder_flag)

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(rating_matrix, dropout_prob)

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

input_type = 2

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

reparameterize(mu, logvar)

training: bool

update_prior()

recbole.model.general_recommender.recvae.log_norm_pdf(x, mu, logvar)

recbole.model.general_recommender.recvae.swish(x)

Swish activation function:

\[\text{Swish}(x) = \frac{x}{1 + \exp(-x)}\]