TransformerBackboneBase

class rectools.models.nn.transformers.torch_backbone.TransformerBackboneBase(n_heads: int, dropout_rate: float, item_model: ItemNetBase, pos_encoding_layer: PositionalEncodingBase, transformer_layers: TransformerLayersBase, similarity_module: SimilarityModuleBase, use_causal_attn: bool = True, use_key_padding_mask: bool = False, **kwargs: Any)

Bases: Module

Base class for transformer torch backbone.

Methods

encode_sessions(batch, item_embs)	Pass user history through item embeddings.
forward(batch[, candidate_item_ids])	Forward pass to get item and session embeddings.

Attributes

Parameters

• n_heads (int) –

• dropout_rate (float) –

• item_model (ItemNetBase) –

• pos_encoding_layer (PositionalEncodingBase) –

• transformer_layers (TransformerLayersBase) –

• similarity_module (SimilarityModuleBase) –

• use_causal_attn (bool) –

• use_key_padding_mask (bool) –

• kwargs (Any) –

encode_sessions(batch: Dict[str, Tensor], item_embs: Tensor) → Tensor

Pass user history through item embeddings. Add positional encoding. Pass history through transformer blocks.

Parameters

• batch (Dict[str, torch.Tensor]) – Dictionary containing user sessions data.

• item_embs (torch.Tensor) – Item embeddings.

Returns

Encoded session embeddings.

Return type

torch.Tensor. [batch_size, session_max_len, n_factors]

forward(batch: Dict[str, Tensor], candidate_item_ids: Optional[Tensor] = None) → Tensor

Forward pass to get item and session embeddings. Get item embeddings. Pass user sessions through transformer blocks.

Parameters

• batch (Dict[str, torch.Tensor]) – Dictionary containing user sessions data, with “x” key containing session tensor.

• candidate_item_ids (optional(torch.Tensor), default None) – Defined item ids for similarity calculation.

Return type

torch.Tensor