如何解释TensorFlow的卷积滤波器和步幅参数？ [英] How to interpret TensorFlow's convolution filter and striding parameters?

问题描述

我试图了解TensorFlow的卷积，尤其是公式

I'm trying to understand TensorFlow's convolution, in particular the formula

shape(output) = [batch,
             (in_height - filter_height + 1) / strides[1],
             (in_width - filter_width + 1) / strides[2],
             ...]

我希望公式为

shape(output) = [batch,
             (in_height - filter_height) / strides[1] + 1,
             (in_width - filter_width) / strides[2] + 1,
             ...]

。从32x32图像开始，并应用带有步幅[1,3,3,1]的5x5滤镜，然后以我的理解，这应该会产生10x10的输出，其值是面积的卷积

instead. Starting from a 32x32 image, and applying a 5x5 filter with strides [1,3,3,1], then in my understanding this should yield a 10x10 output, whose values are the convolutions of the areas

 (0:4,0:4) ,  (0:4,3:7) ,  (0:4,6:10) , ...,  (0:4,27:31), 
 (3:7,0:4) ,  (3:7,3:7) ,  (3:7,6:10) , ...,  (3:7,27:31),
...
(27:31,0:4), (27:31,3:7), (27:31,6:10), ..., (27:31,27:31)

所以两个尺寸都应为地板（（32 -5）/ 3）+ 1 = 10，而不是floor（（32-5 + 1）/ 3）= 9。我在这里想念什么？我是否误解了这里的卷积方式和/或参数的含义？如果是这样，我应该使用什么参数来获得上述选择？

so both dimensions should be floor((32-5)/3)+1=10 and not floor((32-5+1)/3)=9. What am I missing here? Have I misunderstood the way convolution is done here and/or what the parameters mean? If so, what parameters should I use in order to obtain the above selection?

推荐答案

根据问题＃196 ，这部分文档显然是错误的；我认为dga的答案仍然存在问题。

According to issue #196, this part of the documentation is apparently wrong; and I think there is still problem in dga's answer.

应为：

floor（（in_height + y_padding-filter_height）/ y_stride）+ 1，

floor((in_height+y_padding-filter_height)/y_stride) + 1,

• 当padding = VALID时，y_padding = 0。


• 当padding = SAME时，通常y_padding进行调整以使（in_height + y_padding-filter_height）/ y_stride为整数，从而不必使用地板。

这篇关于如何解释TensorFlow的卷积滤波器和步幅参数？的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！