Python搭建NLP模型的步骤和代码详解

pip install numpy pandas scikit-learn nltk spacy transformers torch tensorflow

import pandas as pd
from sklearn.model_selection import train_test_split

# 读取 IMDB 数据集
url = "https://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz"
df = pd.read_csv("IMDB Dataset.csv")  # 数据集需要提前下载并存储

# 划分数据集
train_texts, test_texts, train_labels, test_labels = train_test_split(df['review'], df['sentiment'], test_size=0.2, random_state=42)

# 转换标签为数值
train_labels = train_labels.map({'positive': 1, 'negative': 0})
test_labels = test_labels.map({'positive': 1, 'negative': 0})

import re
import nltk
from nltk.corpus import stopwords
from nltk.tokenize import word_tokenize

nltk.download('stopwords')
nltk.download('punkt')

# 定义文本清理函数
def clean_text(text):
    text = re.sub(r'<.*?>', '', text)  # 移除 HTML 标签
    text = re.sub(r'[^a-zA-Z]', ' ', text)  # 只保留字母
    tokens = word_tokenize(text.lower())  # 分词
    tokens = [word for word in tokens if word not in stopwords.words('english')]  # 去停用词
    return ' '.join(tokens)

# 处理数据
train_texts = train_texts.apply(clean_text)
test_texts = test_texts.apply(clean_text)

vectorizer = TfidfVectorizer(max_features=5000)
X_train = vectorizer.fit_transform(train_texts)
X_test = vectorizer.transform(test_texts)

from gensim.models import Word2Vec

sentences = [text.split() for text in train_texts]
word2vec_model = Word2Vec(sentences, vector_size=100, window=5, min_count=5, workers=4)
word2vec_model.save("word2vec.model")

from sklearn.linear_model import LogisticRegression
from sklearn.metrics import accuracy_score

model = LogisticRegression()
model.fit(X_train, train_labels)

# 预测
preds = model.predict(X_test)
print("Logistic Regression Accuracy:", accuracy_score(test_labels, preds))

import torch
import torch.nn as nn
import torch.optim as optim
from torch.utils.data import DataLoader, TensorDataset

# 定义 LSTM 模型
class LSTMModel(nn.Module):
    def __init__(self, vocab_size, embedding_dim, hidden_dim, output_dim):
        super(LSTMModel, self).__init__()
        self.embedding = nn.Embedding(vocab_size, embedding_dim)
        self.lstm = nn.LSTM(embedding_dim, hidden_dim, batch_first=True)
        self.fc = nn.Linear(hidden_dim, output_dim)
    
    def forward(self, x):
        embedded = self.embedding(x)
        _, (hidden, _) = self.lstm(embedded)
        return self.fc(hidden.squeeze(0))

# 超参数
VOCAB_SIZE = 5000
EMBEDDING_DIM = 100
HIDDEN_DIM = 128
OUTPUT_DIM = 1

model = LSTMModel(VOCAB_SIZE, EMBEDDING_DIM, HIDDEN_DIM, OUTPUT_DIM)

# 训练模型
criterion = nn.BCEWithLogitsLoss()
optimizer = optim.Adam(model.parameters(), lr=0.001)

for epoch in range(5):
    model.train()
    optimizer.zero_grad()
    outputs = model(torch.randint(0, VOCAB_SIZE, (len(train_labels), 50)))
    loss = criterion(outputs.squeeze(), torch.tensor(train_labels.values, dtype=torch.float))
    loss.backward()
    optimizer.step()
    print(f"Epoch {epoch+1}, Loss: {loss.item()}")

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

tokenizer = BertTokenizer.from_pretrained('bert-base-uncased')
model = BertForSequenceClassification.from_pretrained('bert-base-uncased')

# Tokenize 数据
train_encodings = tokenizer(list(train_texts), truncation=True, padding=True, max_length=512, return_tensors="pt")
test_encodings = tokenizer(list(test_texts), truncation=True, padding=True, max_length=512, return_tensors="pt")

# 转换为 PyTorch Dataset
class IMDbDataset(torch.utils.data.Dataset):
    def __init__(self, encodings, labels):
        self.encodings = encodings
        self.labels = labels

    def __len__(self):
        return len(self.labels)

    def __getitem__(self, idx):
        item = {key: val[idx] for key, val in self.encodings.items()}
        item["labels"] = torch.tensor(self.labels[idx])
        return item

train_dataset = IMDbDataset(train_encodings, list(train_labels))
test_dataset = IMDbDataset(test_encodings, list(test_labels))

# 训练 BERT
train_loader = DataLoader(train_dataset, batch_size=8, shuffle=True)
optimizer = torch.optim.AdamW(model.parameters(), lr=5e-5)

model.train()
for epoch in range(3):
    for batch in train_loader:
        optimizer.zero_grad()
        outputs = model(**batch)
        loss = outputs.loss
        loss.backward()
        optimizer.step()
        print(f"Epoch {epoch+1}, Loss: {loss.item()}")

from sklearn.metrics import classification_report

model.eval()
preds = []
with torch.no_grad():
    for batch in test_dataset:
        output = model(**batch)
        preds.append(torch.argmax(output.logits, axis=1).numpy())

print(classification_report(test_labels, preds))

from fastapi import FastAPI
import torch

app = FastAPI()

@app.post("/predict/")
def predict(text: str):
    encoding = tokenizer(text, return_tensors="pt", truncation=True, padding=True, max_length=512)
    with torch.no_grad():
        output = model(**encoding)
        pred = torch.argmax(output.logits, axis=1).item()
    return {"sentiment": "positive" if pred == 1 else "negative"}