锦中人工智能助手

# 安装必要的库

# pip install transformers torch

from transformers import AutoTokenizer, AutoModelForSequenceClassification

import torch

# 加载预训练模型和分词器

model_name = "bert-base-uncased"

tokenizer = AutoTokenizer.from_pretrained(model_name)

model = AutoModelForSequenceClassification.from_pretrained(model_name)

# 示例数据

texts = ["学生学习时间较长", "课程内容较难"]

labels = [1, 0] # 1表示需要关注，0表示正常

# 对文本进行编码

inputs = tokenizer(texts, padding=True, truncation=True, return_tensors="pt")

# 模型训练

optimizer = torch.optim.AdamW(model.parameters(), lr=5e-5)

for epoch in range(3): # 训练3个周期

model.train()

outputs = model(**inputs, labels=torch.tensor(labels))

loss = outputs.loss

loss.backward()

optimizer.step()

optimizer.zero_grad()

print("训练完成！")

from fastapi import FastAPI

from pydantic import BaseModel

import torch

from transformers import AutoTokenizer, AutoModelForSequenceClassification

app = FastAPI()

# 加载模型和分词器

model_name = "bert-base-uncased"

tokenizer = AutoTokenizer.from_pretrained(model_name)

model = AutoModelForSequenceClassification.from_pretrained("path_to_trained_model")

class PredictionRequest(BaseModel):

text: str

@app.post("/predict")

def predict(request: PredictionRequest):

inputs = tokenizer(request.text, return_tensors="pt")

with torch.no_grad():

outputs = model(**inputs)

prediction = torch.softmax(outputs.logits, dim=1).tolist()[0]

return {"prediction": prediction}