Class AbsolutePositionalEncoding

Models/AbsolutePositionalEncoding.py:47–83

r"""Inject some information about the relative or absolute position of the tokens in the sequence. The positional encodings have the same dimension as the embeddings, so that the two can be summed. Here, we use sine and cosine functions of different frequencies. .. math::

class AbsolutePositionalEncoding(nn.Module):
    r"""Inject some information about the relative or absolute position of the tokens
        in the sequence. The positional encodings have the same dimension as
        the embeddings, so that the two can be summed. Here, we use sine and cosine
        functions of different frequencies.
    .. math::
        \text{PosEncoder}(pos, 2i) = sin(pos/10000^(2i/d_model))
        \text{PosEncoder}(pos, 2i+1) = cos(pos/10000^(2i/d_model))
        \text{where pos is the word position and i is the embed idx)
    Args:
        d_model: the embed dim (required).
        dropout: the dropout value (default=0.1).
        max_len: the max. length of the incoming sequence (default=1024).
    """

    def __init__(self, d_model, dropout=0.1, max_len=1024, scale_factor=1.0):
        super(AbsolutePositionalEncoding, self).__init__()
        self.dropout = nn.Dropout(p=dropout)
        pe = torch.zeros(max_len, d_model)  # positional encoding
        position = torch.arange(0, max_len, dtype=torch.float).unsqueeze(1)
        div_term = torch.exp(torch.arange(0, d_model, 2).float() * (-math.log(10000.0) / d_model))

        pe[:, 0::2] = torch.sin(position * div_term)
        pe[:, 1::2] = torch.cos(position * div_term)
        pe = scale_factor * pe.unsqueeze(0)
        self.register_buffer('pe', pe)  # this stores the variable in the state_dict (used for non-trainable variables)

    def forward(self, x):
        r"""Inputs of forward function
        Args:
            x: the sequence fed to the positional encoder model (required).
        Shape:
            x: [sequence length, batch size, embed dim]
            output: [sequence length, batch size, embed dim]
        """
        x = x + self.pe
        return self.dropout(x)

class LearnablePositionalEncoding(nn.Module):