Keras_Classifier

1 data_pre_procession

2 .model = Sequencial([dense(),relu(),dense(),softmax])

3.可选：自定义优化器，optimizer

4.model.compile(loss='categorical_crossentropy',metrics=['accuracy'],optimizer)

5.model.fit(data_x,data_y,epoch=,metrics=['accuracy'])

6. loss,accuracy = model.evaluate(x_test,y_test)

# 5 - Classifier example

import numpy as np
from keras import optimizers

np.random.seed(1337)
from keras.datasets import mnist
from keras.utils import np_utils
from keras.models import Sequential
from keras.layers import Dense, Activation


# import numpy as np
# data
# download mnist to the path '~/.keras/datasets/' if it is first time to be called
# X shape (60000,28*28),Y_shape(10000)

#data
(X_train, y_train), (X_test, y_test) = mnist.load_data()

print(X_train.shape)  # (60000, 28, 28)
print(X_test.shape)  # (10000, 28, 28)

print(y_train.shape)  # (60000,)
print(y_test.shape)  # (10000,)



#data
#input_reshape
X_train =X_train.reshape(X_train.shape[0],-1) / 255
X_test =X_test.reshape(X_test.shape[0],-1) / 255
#y_data标签预处理
y_train = np_utils.to_categorical(y_train,num_classes=10)
y_test = np_utils.to_categorical(y_test,num_classes = 10)

#add_layers,新建一个序列对象，供下面的编译时候指定
model = Sequential(
    [
        Dense(32,input_dim=784),
        Activation('relu'),
        Dense(10),
        Activation('softmax'),
    ]
)

#可以在编译之前自定义optimizer
rm_sprop = optimizers.RMSprop(lr=0.001, rho=0.9, epsilon=1e-08, decay=0.0)
#compile
model.compile(optimizer=rm_sprop,loss='categorical_crossentropy',metrics=['accuracy'])

#fit
model.fit(X_train,y_train,epochs=5,batch_size=32)#batch
#evaluate
loss,accuracy = model.evaluate(X_test, y_test)

print('test loss:',loss)
print('test accuracy:',accuracy)