目录

问题描述：

代码实现：

1.引入依赖，加载数据

2.数据处理和数据编码

3.构建网络

4.编译模型

5.数据可视化

代码展示：

实现截图：

训练过程

可视化

参考：

问题描述：

使用keras中的顺序模型来分类keras电影评论数据集的二分类问题

代码实现：

1.引入依赖，加载数据

from cProfile import label
from keras.datasets import imdb
import numpy as np
from keras import models
from keras import layers
from keras import optimizers,optimizer_v1,optimizer_v2
import matplotlib.pyplot as plt
from keras import  losses
from keras import metrics


#仅保留数据中前10000单词
(train_data,train_labels),(test_data,test_labels) = imdb.load_data(num_words=10000)

2.数据处理和数据编码

word_index = imdb.get_word_index()
#键值颠倒，将整数索引映射为单词
reverse_word_index = dict(
    [(value,key) for (key,value) in word_index.items()]
)
decoded_review = ''.join([reverse_word_index.get(i-3,'?')for i in train_data[0]])


#编码成为二进制矩阵
def vectorize_sequences(sequences,dimension=10000):
    results = np.zeros((len(sequences),dimension))
    for i ,sequence in enumerate(sequences):
        results[i,sequence]  = 1
    return results


x_train = vectorize_sequences(train_data)
x_test = vectorize_sequences(test_data)

#标签向量化
y_train = np.asarray(train_labels).astype('float32')
y_test = np.asarray(test_labels).astype('float32')

3.构建网络

'''
参考
1.#编译模型
#bianary_crossentropy----二元交叉熵
model.compile(optimizer='rmsprop',loss="binary_crossentropy",metrics=['accuracy'])


2.#配置优化器
model.compile(optimizer=optimizer_v1.RMSprop(lr = 0.001),loss = 'binary_crossentropy',metrics=['accuracy'])

3.#使用自定义得损失和指标
model.compile(optimizer=optimizer_v1.RMSprop(lr = 0.001),loss = losses.binary_crossentropy,metrics=[metrics.binary_accuracy])
'''

#构建为网络
model = models.Sequential()
model.add(layers.Dense(16,activation='relu',input_shape = (10000,)))
model.add(layers.Dense(16,activation='relu'))
model.add(layers.Dense(1,activation='sigmoid'))

4.编译模型

#留出验证集
x_val = x_train[:10000]
partial_x_train = x_train[10000:]

y_val = y_train[:10000]
partial_y_train = y_train[10000:]

#训练模型
model.compile(optimizer='rmsprop',loss="binary_crossentropy",metrics=['acc'])

5.数据可视化

#    1.绘制训练损失和验证损失
history_dict = history.history
loss_value = history_dict['loss']
val_loss_values = history_dict['val_loss']
epochs = range(1,len(loss_value) +1)


plt.plot(epochs,loss_value,'bo',label='Training loss')
plt.plot(epochs,val_loss_values,'b',label="validation loss")
plt.title('Training and validation loss')
plt.xlabel('Epochs')
plt.ylabel("loss")
plt.legend()
plt.show()


#2.    绘制训练精度和验证精度
plt.clf()#清空图像
acc = history_dict['acc']
val_acc = history_dict['val_acc']
plt.plot(epochs,acc,'bo',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.show()

代码展示：

from keras.datasets import imdb
import numpy as np
from keras import models
from keras import layers
from keras import optimizers,optimizer_v1,optimizer_v2
import matplotlib.pyplot as plt
from keras import  losses
from keras import metrics


#仅保留数据中前10000单词
(train_data,train_labels),(test_data,test_labels) = imdb.load_data(num_words=10000)


word_index = imdb.get_word_index()
#键值颠倒，将整数索引映射为单词
reverse_word_index = dict(
    [(value,key) for (key,value) in word_index.items()]
)
decoded_review = ''.join([reverse_word_index.get(i-3,'?')for i in train_data[0]])


#编码成为二进制矩阵
def vectorize_sequences(sequences,dimension=10000):
    results = np.zeros((len(sequences),dimension))
    for i ,sequence in enumerate(sequences):
        results[i,sequence]  = 1
    return results


x_train = vectorize_sequences(train_data)
x_test = vectorize_sequences(test_data)

#标签向量化
y_train = np.asarray(train_labels).astype('float32')
y_test = np.asarray(test_labels).astype('float32')

#构建为网络
model = models.Sequential()
model.add(layers.Dense(16,activation='relu',input_shape = (10000,)))
model.add(layers.Dense(16,activation='relu'))
model.add(layers.Dense(1,activation='sigmoid'))


#留出验证集
x_val = x_train[:10000]
partial_x_train = x_train[10000:]

y_val = y_train[:10000]
partial_y_train = y_train[10000:]

#训练模型
model.compile(optimizer='rmsprop',loss="binary_crossentropy",metrics=['acc'])

'''
history_dict.keys()
>>>dict_keys(['val_acc', 'acc', 'val_loss', 'loss'])
'''
history = model.fit(partial_x_train,partial_y_train,epochs = 20,batch_size =512,validation_data=(x_val,y_val))


#绘制训练损失和验证损失
history_dict = history.history
loss_value = history_dict['loss']
val_loss_values = history_dict['val_loss']
epochs = range(1,len(loss_value) +1)


plt.plot(epochs,loss_value,'bo',label='Training loss')
plt.plot(epochs,val_loss_values,'b',label="validation loss")
plt.title('Training and validation loss')
plt.xlabel('Epochs')
plt.ylabel("loss")
plt.legend()
plt.show()


#绘制训练精度和验证精度
plt.clf()#清空图像
acc = history_dict['acc']
val_acc = history_dict['val_acc']
plt.plot(epochs,acc,'bo',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.show()

实现截图：

训练过程

可视化

参考：

《Python深度学习》