python人工智能 NLP BERT 文本分类 : 用 BERT 实现一个 NLP 文本分类 任务，代码方案分享1(图文详解)

import torch
from torch.utils.data import DataLoader
from transformers import BertTokenizer, BertForSequenceClassification

# 加载数据
train_data = ...
test_data = ...

# 初始化分词器和模型
tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
model = BertForSequenceClassification.from_pretrained('bert-base-uncased')

# 创建数据加载器
train_dataloader = DataLoader(train_data, batch_size=32)
test_dataloader = DataLoader(test_data, batch_size=32)

# 定义优化器和损失函数
optimizer = torch.optim.AdamW(model.parameters(), lr=2e-5)
loss_fn = torch.nn.CrossEntropyLoss()

# 训练模型
for epoch in range(3):
    for batch in train_dataloader:
        inputs = tokenizer(batch['text'], return_tensors="pt")
        outputs = model(**inputs)
        loss = loss_fn(outputs.logits, batch['label'])
        loss.backward()
        optimizer.step()
        optimizer.zero_grad()

# 评估模型
with torch.no_grad():
    for batch in test_dataloader:
        inputs = tokenizer(batch['text'], return_tensors="pt")
        outputs = model(**inputs)
        predictions = torch.argmax(outputs.logits, dim=-1)
        accuracy = (predictions == batch['label']).float().mean()
        print(f"Accuracy: {accuracy}")