使用RNN实现自然语言词性标注代码

以下是使用RNN实现自然语言词性标注的示例代码，基于PyTorch框架：

import torch
import torch.nn as nn
import torch.optim as optim

# 定义模型类
class RNN_POS(nn.Module):
    def __init__(self, vocab_size, embedding_dim, hidden_dim, output_size):
        super(RNN_POS, self).__init__()
        self.embedding = nn.Embedding(vocab_size, embedding_dim)
        self.rnn = nn.RNN(embedding_dim, hidden_dim, batch_first=True)
        self.fc = nn.Linear(hidden_dim, output_size)
    
    def forward(self, x):
        x = self.embedding(x)
        x, _ = self.rnn(x)
        x = self.fc(x)
        return x

# 定义训练函数
def train(model, train_loader, optimizer, criterion):
    model.train()
    for batch_idx, (data, target) in enumerate(train_loader):
        optimizer.zero_grad()
        output = model(data)
        loss = criterion(output.view(-1, output.shape[-1]), target.view(-1))
        loss.backward()
        optimizer.step()

# 定义测试函数
def test(model, test_loader, criterion):
    model.eval()
    test_loss = 0
    correct = 0
    with torch.no_grad():
        for data, target in test_loader:
            output = model(data)
            test_loss += criterion(output.view(-1, output.shape[-1]), target.view(-1)).item()
            pred = output.argmax(dim=-1)
            correct += pred.eq(target).sum().item()
    test_loss /= len(test_loader.dataset)
    accuracy = 100. * correct / len(test_loader.dataset)
    print('Test set: Average loss: {:.4f}, Accuracy: {}/{} ({:.0f}%)'.format(
        test_loss, correct, len(test_loader.dataset), accuracy))

# 定义超参数
vocab_size = 10000
embedding_dim = 128
hidden_dim = 64
output_size = 10
lr = 0.01
epochs = 10
batch_size = 64

# 加载数据集（这里使用PyTorch自带的Penn Treebank数据集）
from torchtext.datasets import PennTreebank
from torchtext.data.utils import get_tokenizer
from torchtext.data.utils import ngrams_iterator
from torchtext.vocab import build_vocab_from_iterator

tokenizer = get_tokenizer('basic_english')
train_iter = PennTreebank(split='train')
train_text = []
for (text,) in train_iter:
    train_text += tokenizer(text)
vocab = build_vocab_from_iterator(ngrams_iterator(train_text, 1, 2))
vocab.set_default_index(vocab["<unk>"])
train_iter, val_iter, test_iter = PennTreebank()
train_loader = torch.utils.data.DataLoader(train_iter, batch_size=batch_size, shuffle=True)
test_loader = torch.utils.data.DataLoader(test_iter, batch_size=batch_size, shuffle=True)

# 初始化模型、优化器、损失函数
model = RNN_POS(vocab_size, embedding_dim, hidden_dim, output_size)
optimizer = optim.Adam(model.parameters(), lr=lr)
criterion = nn.CrossEntropyLoss()

# 训练模型
for epoch in range(1, epochs+1):
    train(model, train_loader, optimizer, criterion)
    test(model, test_loader, criterion)

这个示例代码使用Penn Treebank数据集进行训练和测试，输出结果包括测试集的平均损失和准确率。你可以将其修改为适用于其他数据集的代码。