transformer模型分类器

Transformer模型可以用作分类器，它在NLP任务中表现出色。下面是一个使用Transformer模型进行文本分类的示例[^2]：

import torch
import torch.nn as nn
import torch.optim as optim
from torchtext.data import Field, TabularDataset, BucketIterator
from torch.nn import TransformerEncoder, TransformerEncoderLayer

# 定义模型
class TransformerClassifier(nn.Module):
    def __init__(self, input_dim, hidden_dim, output_dim, n_layers, n_heads, dropout):
        super().__init__()
        self.embedding = nn.Embedding(input_dim, hidden_dim)
        self.encoder_layer = TransformerEncoderLayer(hidden_dim, n_heads, hidden_dim, dropout)
        self.encoder = TransformerEncoder(self.encoder_layer, n_layers)
        self.fc = nn.Linear(hidden_dim, output_dim)
        self.dropout = nn.Dropout(dropout)
        
    def forward(self, text):
        embedded = self.embedding(text)
        embedded = embedded.permute(1, 0, 2)
        output = self.encoder(embedded)
        output = output.permute(1, 0, 2)
        pooled = torch.mean(output, dim=1)
        pooled = self.dropout(pooled)
        return self.fc(pooled)

# 定义数据集和字段
TEXT = Field(tokenize='spacy', lower=True)
LABEL = Field(sequential=False, is_target=True)

fields = [('text', TEXT), ('label', LABEL)]

# 加载数据集
train_data, test_data = TabularDataset.splits(
    path='data',
    train='train.csv',
    test='test.csv',
    format='csv',
    fields=fields,
    skip_header=True
)

# 构建词汇表
TEXT.build_vocab(train_data, min_freq=2)
LABEL.build_vocab(train_data)

# 创建迭代器
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
train_iterator, test_iterator = BucketIterator.splits(
    (train_data, test_data),
    batch_size=64,
    sort_key=lambda x: len(x.text),
    sort_within_batch=True,
    device=device
)

# 初始化模型和优化器
input_dim = len(TEXT.vocab)
output_dim = len(LABEL.vocab)
hidden_dim = 256
n_layers = 2
n_heads = 4
dropout = 0.5

model = TransformerClassifier(input_dim, hidden_dim, output_dim, n_layers, n_heads, dropout).to(device)
optimizer = optim.Adam(model.parameters())

# 训练模型
def train(model, iterator, optimizer, criterion):
    model.train()
    epoch_loss = 0
    for batch in iterator:
        optimizer.zero_grad()
        text = batch.text
        label = batch.label
        predictions = model(text).squeeze(1)
        loss = criterion(predictions, label)
        loss.backward()
        optimizer.step()
        epoch_loss += loss.item()
    return epoch_loss / len(iterator)

# 测试模型
def evaluate(model, iterator, criterion):
    model.eval()
    epoch_loss = 0
    with torch.no_grad():
        for batch in iterator:
            text = batch.text
            label = batch.label
            predictions = model(text).squeeze(1)
            loss = criterion(predictions, label)
            epoch_loss += loss.item()
    return epoch_loss / len(iterator)

# 训练和测试模型
N_EPOCHS = 10
criterion = nn.CrossEntropyLoss().to(device)

for epoch in range(N_EPOCHS):
    train_loss = train(model, train_iterator, optimizer, criterion)
    test_loss = evaluate(model, test_iterator, criterion)
    print(f'Epoch: {epoch+1}, Train Loss: {train_loss:.4f}, Test Loss: {test_loss:.4f}')

# 使用模型进行预测
def predict_sentiment(model, sentence):
    model.eval()
    tokenized = [tok.text for tok in nlp.tokenizer(sentence)]
    indexed = [TEXT.vocab.stoi[t] for t in tokenized]
    tensor = torch.LongTensor(indexed).unsqueeze(1).to(device)
    prediction = torch.argmax(model(tensor))
    return LABEL.vocab.itos[prediction]

# 示例预测
example_sentence = "This movie is great!"
predicted_label = predict_sentiment(model, example_sentence)
print(f'Predicted Label: {predicted_label}')

这是一个基本的Transformer模型分类器的示例。你可以根据自己的数据集和任务进行相应的修改和调整。