fconv.py 文件源码

def __init__(self, dictionary, embed_dim=512, max_positions=1024,
                 convolutions=((512, 3),) * 20, dropout=0.1):
        super().__init__()
        self.dictionary = dictionary
        self.dropout = dropout
        self.num_attention_layers = None

        num_embeddings = len(dictionary)
        padding_idx = dictionary.pad()
        self.embed_tokens = Embedding(num_embeddings, embed_dim, padding_idx)
        self.embed_positions = Embedding(max_positions, embed_dim, padding_idx)

        in_channels = convolutions[0][0]
        self.fc1 = Linear(embed_dim, in_channels, dropout=dropout)
        self.projections = nn.ModuleList()
        self.convolutions = nn.ModuleList()
        for (out_channels, kernel_size) in convolutions:
            pad = (kernel_size - 1) / 2
            self.projections.append(Linear(in_channels, out_channels)
                                    if in_channels != out_channels else None)
            self.convolutions.append(
                ConvTBC(in_channels, out_channels * 2, kernel_size, padding=pad,
                        dropout=dropout))
            in_channels = out_channels
        self.fc2 = Linear(in_channels, embed_dim)