现有数据集敏感词的训练集和测试集txt文本，其中训练集和测试集分别有非敏感词txt文本和敏感词txt文本，txt文本中文字的格式为，每一个词语单独成行，使用python，CNN算法进行机器学习训练，使得在测试集上表现良好，给出代码，给出f1值的计算auc——roc曲线的绘制

在Python中使用CNN (卷积神经网络) 对敏感词检测的数据集进行训练，首先需要安装必要的库如`torch`, `numpy`, `pandas`, `sklearn`等。这里是一个简单的步骤概述：

1. **数据预处理**:
- 导入所需的库并读取txt文件到DataFrame。

   import pandas as pd
   from sklearn.model_selection import train_test_split

   def load_data(file_path):
       df = pd.read_csv(file_path, sep='\n', header=None)
       return df.values

   X_train, X_test = train_test_split(load_data('train.txt'), test_size=0.2)
   y_train, y_test = load_data('labels_train.txt'), load_data('labels_test.txt')

2. **文本向量化**:
使用`Keras`的`Tokenizer`将文本转换为数字序列。

   from tensorflow.keras.preprocessing.text import Tokenizer
   from tensorflow.keras.preprocessing.sequence import pad_sequences

   tokenizer = Tokenizer()
   tokenizer.fit_on_texts(X_train)
   X_train_seq = tokenizer.texts_to_sequences(X_train)
   X_test_seq = tokenizer.texts_to_sequences(X_test)

   max_len = max(len(seq) for seq in X_train_seq)
   X_train_padded = pad_sequences(X_train_seq, maxlen=max_len)
   X_test_padded = pad_sequences(X_test_seq, maxlen=max_len)

3. **构建CNN模型**:

   from tensorflow.keras.models import Sequential
   from tensorflow.keras.layers import Embedding, Conv1D, GlobalMaxPooling1D, Dense

   model = Sequential([
       Embedding(tokenizer.num_words, 16, input_length=max_len),
       Conv1D(32, kernel_size=3, activation='relu'),
       GlobalMaxPooling1D(),
       Dense(64, activation='relu'),
       Dense(1, activation='sigmoid')  # binary classification with sigmoid output
   ])

   model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])

4. **训练模型**:

   history = model.fit(X_train_padded, y_train, validation_data=(X_test_padded, y_test), epochs=10)

5. **评估性能**:

   y_pred = model.predict(X_test_padded)
   f1_score = f1_score(y_test, y_pred.round())
   
   from sklearn.metrics import roc_auc_score, roc_curve
   auc = roc_auc_score(y_test, y_pred)
   fpr, tpr, _ = roc_curve(y_test, y_pred)

   plt.figure()
   lw = 2
   plt.plot(fpr, tpr, color='darkorange',
            lw=lw, label=f'ROC curve (area = {auc:.2f})')
   plt.plot([0, 1], [0, 1], color='navy', lw=lw, linestyle='--')
   plt.xlim([0.0, 1.0])
   plt.ylim([0.0, 1.05])
   plt.xlabel('False Positive Rate')
   plt.ylabel('True Positive Rate')
   plt.title('Receiver Operating Characteristic')
   plt.legend(loc="lower right")
   plt.show()

这里假设`y_test`已经被转换成了二进制标签，以便于计算F1分数和绘制AUC-ROC曲线。