去掉CNN层遇到的一些小白问题,希望得到大佬的解答

[nlper01]

大佬，我想去掉整个CNN层，即三个Embeding层出来后直接接到协预测器模块，在修改过程遇到一些问题，一下没法解决了，希望得到大佬的帮助。 注释掉送入卷积层的channel

      conv_input_size = config.lstm_hid_size + config.dist_emb_size + config.type_emb_size
        self.convLayer = ConvolutionLayer(conv_input_size, config.conv_hid_size, config.dilation, config.conv_dropout)
        self.dropout = nn.Dropout(config.emb_dropout)

并把这里

 conv_inputs = torch.cat([dis_emb, reg_emb, cln], dim=-1)
        conv_inputs = torch.masked_fill(conv_inputs, grid_mask2d.eq(0).unsqueeze(-1), 0.0)
        conv_outputs = self.convLayer(conv_inputs)
        conv_outputs = torch.masked_fill(conv_outputs, grid_mask2d.eq(0).unsqueeze(-1), 0.0)
        outputs = self.predictor(word_reps, word_reps, conv_outputs)

修改为

conv_inputs = torch.cat([dis_emb, reg_emb, cln], dim=-1)
       conv_outputs = torch.masked_fill(conv_inputs, grid_mask2d.eq(0).unsqueeze(-1), 0.0)
       outputs = self.predictor(word_reps, word_reps, conv_outputs)

然后以下地方报错说 channels 未定义，现在还未解决，不知道像我这样操作，config.conv_hid_size * len(config.dilation)需要改成什么？dis_embs+lstm_hidesize+reg_embs?

self.predictor = CoPredictor(config.label_num, config.lstm_hid_size, config.biaffine_size,
                                    config.conv_hid_size * len(config.dilation), config.ffnn_hid_size,
                                     config.out_dropout

class CoPredictor(nn.Module):
    def __init__(self, cls_num, hid_size, biaffine_size, channels, ffnn_hid_size, dropout=0):
        super().__init__()
        self.mlp1 = MLP(n_in=hid_size, n_out=biaffine_size, dropout=dropout)
        self.mlp2 = MLP(n_in=hid_size, n_out=biaffine_size, dropout=dropout)
        self.biaffine = Biaffine(n_in=biaffine_size, n_out=cls_num, bias_x=True, bias_y=True)
       self.mlp_rel = MLP(channels, ffnn_hid_size, dropout=dropout)
        self.linear = nn.Linear(ffnn_hid_size, cls_num)
        self.dropout = nn.Dropout(dropout)

[ljynlp]

用conv_input_size替换掉config.conv_hid_size * len(config.dilation)即可

[nlper01]

用conv_input_size替换掉config.conv_hid_size * len(config.dilation)即可

好的，谢谢