从图片到x_train,y_train的过程汇总

1、这个是从.npz 到x_train，y_train，至于从jpg到npz以后再总结：

(x_tain,y_train),(x_test,y_test) = mnist.loda_data('mnist.npz')
if K.image_data_format() == 'channels first':
	x_train = x_train.reshape([batch_size,1,img_rows,img_cols])  #step 1
	x_test = x_test.reshape([batch_size,1,img_rows,img_cols])
	input_shape = [1,img_rows,img_cols]
else:
	x_train = x_train.reshape([batch_size,img_rows,img_cols,1])
	x_test = x_test.reshape([batch_size,img_rows,img_cols,1])
	input_shape = [img_rows,img_cols,1]

x_train = x_train.astype('float32')     #step 2  GPU对单精度的计算性能更好
x_test = x_test.astype('float32')     #更正，第一遍写的这两块的先后顺序弄反了

x_train /=255
x_test /=255       #step 3



y_train = keras.utils.to_catagorical(y_train,num_classes)
y_test = keras.utils.to_categorical(y_test,num_classes)

得到了x_train ,y_train 就可以作为model.fit(x_train,y_train )的输入了
逻辑：以模型为中心，首先了解模型需要的输入，Keras 这里的model.fit(）方法 就是需要x_train，y_train，shape 是 [总样本数，img_rows,img_cols,通道数]，注意这个方法里的是总样本数，要实