hopwise.model.sequential_recommender.bert4rec¶

Reference:: Fei Sun et al. “BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer.” In CIKM 2019.
Reference code:: The authors’ tensorflow implementation https://github.com/FeiSun/BERT4Rec

Classes¶

BERT4Rec

This is a abstract sequential recommender. All the sequential model should implement This class.

Module Contents¶

class hopwise.model.sequential_recommender.bert4rec.BERT4Rec(config, dataset)¶

Bases: hopwise.model.abstract_recommender.SequentialRecommender

This is a abstract sequential recommender. All the sequential model should implement This class.

n_layers¶

n_heads¶

hidden_size¶

inner_size¶

hidden_dropout_prob¶

attn_dropout_prob¶

hidden_act¶

layer_norm_eps¶

mask_ratio¶

MASK_ITEM_SEQ¶

POS_ITEMS¶

NEG_ITEMS¶

MASK_INDEX¶

loss_type¶

initializer_range¶

mask_token¶

mask_item_length¶

item_embedding¶

position_embedding¶

trm_encoder¶

LayerNorm¶

dropout¶

output_ffn¶

output_gelu¶

output_ln¶

output_bias¶

_init_weights(module)¶: Initialize the weights

reconstruct_test_data(item_seq, item_seq_len)¶: Add mask token at the last position according to the lengths of item_seq

forward(item_seq)¶

multi_hot_embed(masked_index, max_length)¶

For memory, we only need calculate loss for masked position. Generate a multi-hot vector to indicate the masked position for masked sequence, and then is used for gathering the masked position hidden representation.

Examples

sequence: [1 2 3 4 5]

masked_sequence: [1 mask 3 mask 5]

masked_index: [1, 3]

max_length: 5

multi_hot_embed: [[0 1 0 0 0], [0 0 0 1 0]]

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

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