import tensorflow as tf from tensorflow import keras from tensorflow.keras import layers

import matplotlib.pyplot as plt %matplotlib inline

加载Tensorflow的dataset数据

1
2
3
data = keras.datasets.imdb#电影评论数据
max_word = 10000
(x_train,y_train),(x_test,y_test) = data.load_data(num_words=max_word)

文本训练场密集向量

1
2
x_train = keras.preprocessing.sequence.pad_sequences(x_train,300)
x_test = keras.preprocessing.sequence.pad_sequences(x_test,300)

创建模型

1
2
3
4
5
6
7
model = keras.models.Sequential()
model.add(layers.Embedding(10000,50,input_length=300))
#（None,300,50）
#model.add(layers.Flatten())#Flatten转换为二维
model.add(layers.GlobalAveragePooling1D())#Flatten转换为二维
model.add(layers.Dense(64,activation='relu'))
model.add(layers.Dense(1,activation='sigmoid'))#二分类问题，y_train

配置模型

1
2
3
4
model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=0.001),
             loss=tf.keras.losses.BinaryCrossentropy(),
              metrics=['acc']
)

模型训练

1
history = model.fit(x_train,y_train,epochs=15,batch_size=256,validation_data=(x_test,y_test))

查看history中的参数

1
history.history.keys()

dict_keys(['loss', 'acc', 'val_loss', 'val_acc'])

绘制loss变化曲线

1
2
plt.plot(history.epoch,history.history['loss'],'r')
plt.plot(history.epoch,history.history['val_loss'],'b--')

绘制accuracy变化曲线

1
2
plt.plot(history.epoch,history.history['acc'],'r')
plt.plot(history.epoch,history.history['val_acc'],'b--')

解决过拟合问题：1：dropout层 2：l2，l1