textcnn文本分类代码keras

以下是使用Keras实现TextCNN文本分类的示例代码：

import numpy as np
from keras.models import Model
from keras.layers import Input, Dense, Dropout, Embedding, Conv1D, GlobalMaxPooling1D
from keras.preprocessing.sequence import pad_sequences
from keras.utils import to_categorical
from sklearn.model_selection import train_test_split


# 读取数据
def read_data(filename):
    sentences = []
    labels = []
    with open(filename, 'r', encoding='utf-8') as f:
        for line in f:
            sentence, label = line.strip().split('\t')
            sentences.append(sentence)
            labels.append(label)
    return sentences, labels


# 处理数据
def process_data(sentences, labels, max_len, word_dict):
    # 将文本转换为数字序列
    X = []
    for sentence in sentences:
        words = sentence.split(' ')
        seq = []
        for word in words:
            if word in word_dict:
                seq.append(word_dict[word])
            else:
                seq.append(0)
        X.append(seq)
    # 对数据进行填充和截断
    X = pad_sequences(X, maxlen=max_len, padding='post', truncating='post', value=0)
    # 将标签转换为one-hot编码
    Y = to_categorical(np.array(labels))
    return X, Y


# 构建模型
def build_model(max_len, max_word, embedding_dim):
    # 输入层
    inputs = Input(shape=(max_len,))
    # 词嵌入层
    x = Embedding(max_word, embedding_dim)(inputs)
    # 卷积层和池化层
    x = Conv1D(128, 5, activation='relu')(x)
    x = GlobalMaxPooling1D()(x)
    # 全连接层和输出层
    x = Dense(128, activation='relu')(x)
    x = Dropout(0.5)(x)
    outputs = Dense(2, activation='softmax')(x)
    # 构建模型
    model = Model(inputs=inputs, outputs=outputs)
    model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])
    return model


# 主函数
if __name__ == '__main__':
    # 读取数据
    sentences, labels = read_data('data.txt')
    # 构建词典
    word_dict = {}
    for sentence in sentences:
        for word in sentence.split(' '):
            if word not in word_dict:
                word_dict[word] = len(word_dict) + 1
    # 处理数据
    max_len = 100
    X, Y = process_data(sentences, labels, max_len, word_dict)
    # 划分训练集和测试集
    X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.2, random_state=42)
    # 构建模型
    max_word = len(word_dict) + 1
    embedding_dim = 100
    model = build_model(max_len, max_word, embedding_dim)
    # 训练模型
    model.fit(X_train, Y_train, batch_size=64, epochs=10, validation_data=(X_test, Y_test))

这是一个基本的TextCNN模型，可以根据实际需求进行调整和优化。