如果border_mode选择为same,那么卷积操作的输入和输出尺寸会保持一致。如果选择valid,那卷积过后,尺寸会变小

# apply a 3x3 convolution with 64 output filters on a 256x256 image:model = Sequential()model.add(Convolution2D(64, 3, 3, border_mode='same', input_shape=(3, 256, 256))) # now model.output_shape == (None, 64, 256, 256) >> definition in keras: def conv_output_length(input_length, filter_size, border_mode, stride): if input_length is None: return None assert border_mode in {'same', 'valid'} if border_mode == 'same': output_length = input_length elif border_mode == 'valid': output_length = input_length - filter_size + 1 return (output_length + stride - 1) // stride