AI：神经网络IMDB电影评论二分类模型训练和评估

偏执的太偏执、 2023-06-18 02:58 52阅读 0赞

# AI：Keras神经网络IMDB电影评论二分类模型训练和评估，python #

import keras
    from keras.layers import Dense
    from keras import models
    from keras import optimizers
    from keras.datasets import imdb
    from keras import losses
    import numpy as np
    import matplotlib.pyplot as plt
    
    
    def fig_loss(history):
        history_dict = history.history
        loss_values = history_dict['loss']
        val_loss_values = history_dict['val_loss']
        epochs = range(1, len(loss_values) + 1)
        plt.plot(epochs, loss_values, 'b-.', label='Training loss')
        plt.plot(epochs, val_loss_values, 'r--', label='Validation loss')
        plt.title('Training and validation loss')
        plt.xlabel('Epochs')
        plt.ylabel('Loss')
        plt.legend()
        plt.grid()
        plt.show()
    
    
    def fig_acc(history):
        history_dict = history.history
        acc = history_dict['accuracy']
        val_acc = history_dict['val_accuracy']
        epochs = range(1, len(acc) + 1)
        plt.plot(epochs, acc, 'g:', label='Training acc')
        plt.plot(epochs, val_acc, 'r-', label='Validation acc')
        plt.title('Training and validation accuracy')
        plt.xlabel('Epochs')
        plt.ylabel('Accuracy')
        plt.legend()
        plt.grid()
        plt.show()
    
    
    def to_array(a, col):
        for i in range(len(a)):
            a[i] = keras.utils.to_categorical(np.array(a[i]), num_classes=col)
            a[i] = np.sum(a[i], axis=0)
        return np.array(a.tolist())
    
    
    if __name__ == "__main__":
        SIZE = 10000
    
        model = models.Sequential()
        model.add(Dense(64, activation='relu', input_shape=(SIZE,)))
        model.add(Dense(32, activation='relu'))
        model.add(Dense(1, activation='sigmoid'))
    
        model.compile(optimizer=optimizers.RMSprop(lr=0.001),
                      loss=losses.binary_crossentropy,
                      metrics=['accuracy'])
    
        (train_data, train_labels), (test_data, test_labels) = imdb.load_data(num_words=SIZE)
    
        y_train = np.asarray(train_labels).astype('float32')
    
        x_test = to_array(test_data, SIZE)
        test_labels = np.asarray(test_labels).astype('float32')
    
        x_train = to_array(train_data, SIZE)
    
        x = x_train[:SIZE]
        x_val = x_train[SIZE:]
    
        y = train_labels[:SIZE]
        y_val = train_labels[SIZE:]
    
        history = model.fit(x, y, epochs=8,  validation_data=(x_val, y_val))
    
        results = model.evaluate(x_test, test_labels)
        print('训练结束,评估模型->')
        print('损失:')
        print(results[0])
    
        print('精度:')
        print(results[1])
    
        # 绘制训练损失和验证损失
        fig_loss(history)
    
        # 绘制训练精度和验证精度
        fig_acc(history)

输出：

训练结束,评估模型->
    损失:
    1.0794099688053131
    精度:
    0.8561199903488159

训练损失和验证损失：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70][]

训练精度和验证精度：

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 1][]

附：

《AI：普通列表数组转化为one-hot编码的numpy数组矩阵》

[https://zhangphil.blog.csdn.net/article/details/103315347][https_zhangphil.blog.csdn.net_article_details_103315347]

[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70]: https://img-blog.csdnimg.cn/20191129202505910.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA==,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA_size_16_color_FFFFFF_t_70 1]: https://img-blog.csdnimg.cn/20191129202532344.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly96aGFuZ3BoaWwuYmxvZy5jc2RuLm5ldA==,size_16,color_FFFFFF,t_70
[https_zhangphil.blog.csdn.net_article_details_103315347]: https://zhangphil.blog.csdn.net/article/details/103315347