Keras大坑

持续踩坑：

1. 如果model.fit中的参数suffle=True时，会随机打算每一次epoch的数据。（默认打乱），但是验证数据默认不会打乱。
2. fit函数里，先执行validation_split 再 执行shuffle=True，所以val很有可能全是某一类样本了。
3. 多图像输入的generator写法：

generator = ImageDataGenerator(
    featurewise_center=True,
    featurewise_std_normalization=True,
    rotation_range=20,
    width_shift_range=0.2,
    height_shift_range=0.2,
    horizontal_flip=True)

def generate_data_generator_for_two_images(X1, X2, Y):
    genX1 = generator.flow(X1,Y, seed=7)
    genX2 = generator.flow(X2, seed=7)
    while True:
            X1i = genX1.next()
            X2i = genX2.next()
            yield [X1i[0], X2i ], X1i[1]


model.fit_generator(generate_data_generator_for_two_images(audio, image, label),
                     steps_per_epoch=step, epochs=epochs,verbose=2, callbacks=callbacks_list,
                     validation_data=([audio_val,image_val],val_label),        
                     max_q_size=3*batch_size, workers=4, pickle_safe=False)