keras—神经网络CNN—CIFAR_10图像识别

 1 from keras.datasets import cifar10
 from keras.utils import np_utils
 import matplotlib.pyplot as plt
 from keras.models import load_model
 import numpy as np
 np.random.seed()
 (x_img_train,y_label_train),(x_img_test,y_label_test)=cifar10.load_data()
 print('train data=',len(x_img_train))
 print('test data=',len(x_img_test))
 print('x_train_image:',x_img_train.shape)
 print('x_test_image:',x_img_test.shape)
 x_img_train_4D=x_img_train.reshape(x_img_train.shape[],,,).astype('float32')
 x_img_test_4D=x_img_test.reshape(x_img_test.shape[],,,).astype('float32')
 x_img_train_normalize=x_img_train_4D/255.0
 x_img_test_normalize=x_img_test_4D/255.0
 print(x_img_train_normalize[][][])
 y_label_train_OneHot=np_utils.to_categorical(y_label_train)
 y_label_test_OneHot=np_utils.to_categorical(y_label_test)
 print(y_label_train_OneHot[:])
 from keras.models import Sequential
 from keras.layers import Dense,Dropout,Flatten,Conv2D,MaxPooling2D,ZeroPadding2D,Activation
 model=Sequential()
 model.add(Conv2D(filters=,
                  kernel_size=(,),
                  padding='same',
                  input_shape=(,,),
                  activation='relu'))
 model.add(Dropout(0.25))
 model.add(MaxPooling2D(pool_size=(,)))
 model.add(Conv2D(filters=,
                  kernel_size=(,),
                  padding='same',
                  activation='relu'))
 model.add(Dropout(0.25))
 model.add(MaxPooling2D(pool_size=(,)))
 model.add(Flatten())
 model.add(Dropout(0.25))
 model.add(Dense(,activation='relu'))
 model.add(Dropout(0.25))
 model.add(Dense(,activation='softmax'))
 print(model.summary())
 model.compile(loss='categorical_crossentropy',
               optimizer='adam',metrics=['accuracy'])
 try:
     model=load_model("CnnModel.h5")
     print("Load model successfully!Continuous training model!......")
 except :
     print("Failure of loading model!Start training a new model......")

 train_history=model.fit(x=x_img_train_normalize,
                         y=y_label_train_OneHot,validation_split=0.2,
                         epochs=,batch_size=,verbose=)
 model.save("CnnModel.h5")
 print("Saved model to disk")
 def show_train_history(train_history,train,validation):
     plt.plot(train_history.history[train])
     plt.plot(train_history.history[validation])
     plt.title('Train History')
     plt.ylabel(train)
     plt.xlabel('Epoch')
     plt.legend(['train','validation'],loc='upper left')    #显示左上角标签
     plt.show()
 show_train_history(train_history,'acc','val_acc')
 show_train_history(train_history,'loss','val_loss')
 scores=model.evaluate(x_img_test_normalize,y_label_test_OneHot)
 print()
 print('accuracy',scores[])
 prediction=model.predict_classes(x_img_test_normalize)
 print("prediction[:10]",prediction[:])
 import matplotlib.pyplot as plt
 label_dict={:"airplane",:"automobile",:"bird",:"cat",:"deer",:"dog",:"frog",:"horse",:"ship",:"truck"}
 def plot_image_labels_prediction_1(image,labels,prediction,idx,num=):
     fig=plt.gcf()
     fig.set_size_inches(,)
     if num>:num=
     for i in range(,num):
         ax=plt.subplot(,,i+)
         ax.imshow(image[idx],cmap='binary')
         title=str(i)+','+label_dict[labels[i][]]
         if len(prediction)>:
             title+="=>"+label_dict[prediction[i]]
         ax.set_title(title,fontsize=)
         ax.set_xticks([]);ax.set_yticks([])
         idx+=
     plt.show()
 plot_image_labels_prediction_1(x_img_test,y_label_test,prediction,,)
 Predicted_Probability=model.predict(x_img_test_normalize)
 def show_Predicted_Probability(y,prediction,x_img_test,Predicted_Probability,i):
     print('label:',label_dict[y[i][]],'predict:',label_dict[prediction[i]])
     plt.figure(figsize=(,))
     plt.imshow(np.reshape(x_img_test[i],(,,)))
     plt.show()
     for j in range():
         print(label_dict[j]+
               'Probability:%1.9f'%(Predicted_Probability[i][j]))
 show_Predicted_Probability(y_label_test,prediction,x_img_test,Predicted_Probability,)
 show_Predicted_Probability(y_label_test,prediction,x_img_test,Predicted_Probability,)

################################