TextCNN模型代码实现

以下是基于PyTorch框架实现的TextCNN模型代码，用于文本分类：

import torch
import torch.nn as nn
import torch.nn.functional as F

class TextCNN(nn.Module):
    def __init__(self, vocab_size, embedding_dim, num_classes, num_filters, filter_sizes, dropout_prob):
        super(TextCNN, self).__init__()
        self.embedding = nn.Embedding(vocab_size, embedding_dim)
        self.convs = nn.ModuleList([
            nn.Conv2d(in_channels=1, out_channels=num_filters, kernel_size=(fs, embedding_dim)) for fs in filter_sizes
        ])
        self.dropout = nn.Dropout(dropout_prob)
        self.fc = nn.Linear(num_filters * len(filter_sizes), num_classes)

    def forward(self, x):
        x = self.embedding(x)  # (batch_size, seq_len, embedding_dim)
        x = x.unsqueeze(1)  # (batch_size, 1, seq_len, embedding_dim)
        x = [F.relu(conv(x)).squeeze(3) for conv in self.convs]  # [(batch_size, num_filters, seq_len - filter_size + 1), ...]
        x = [F.max_pool1d(conv, conv.size(2)).squeeze(2) for conv in x]  # [(batch_size, num_filters), ...]
        x = torch.cat(x, 1)  # (batch_size, num_filters * len(filter_sizes))
        x = self.dropout(x)
        logits = self.fc(x)
        return logits

其中，`vocab_size`表示词汇表大小，`embedding_dim`表示词向量维度，`num_classes`表示分类数量，`num_filters`表示卷积核数量，`filter_sizes`表示卷积核尺寸列表，`dropout_prob`表示dropout概率。在`forward`函数中，先使用`embedding`层将输入的词id转换为词向量，然后进行卷积和池化操作，最后通过全连接层输出分类结果。