Simple GAT layer, similar to https://arxiv.org/abs/1710.10903

Transformer response rapper. Pushes input through transformer and MLP

Simple GCN layer, similar to https://arxiv.org/abs/1609.02907

Sparse version GAT layer, similar to https://arxiv.org/abs/1710.10903

Transformer Decoder layer. :param int n_heads: the number of multihead attention heads. :param int n_layers: number of transformer l

Transformer Decoder layer. :param int n_heads: the number of multihead attention heads. :param int n_layers: number of transformer l

Transformer encoder module. :param int n_heads: the number of multihead attention heads. :param int n_layers: number of transformer

Transformer encoder module. :param int n_heads: the number of multihead attention heads. :param int n_layers: number of transformer

Transformer encoder module. :param int n_heads: the number of multihead attention heads. :param int n_layers: number of transformer

Special function for only sparse region backpropataion layer.

This Interface expects you to implement model with the following reqs: :attribute model.encoder: takes input returns tuple (enc_

Model which takes context, memories, candidates and encodes them