# Question about group convolution

**URL:** <https://discuss.pytorch.org/t/question-about-group-convolution/17042>\
**Category:** Uncategorized\
**Created:** [April 26, 2018, 10:05am UTC](https://discuss.pytorch.org/t/question-about-group-convolution/17042 "2018-04-26T10:05:43Z")\
**Posts on this page:** 1\
**Showing post:** 12

<div class="post-metadata">

**Author:** ![malcolmyanguci](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/m/cab0a1/32.png) [@malcolmyanguci](https://discuss.pytorch.org/u/malcolmyanguci)\
**Post date:** [May 16, 2020, 10:28am UTC](https://discuss.pytorch.org/t/question-about-group-convolution/17042/12 "2020-05-16T10:28:44Z")

</div>

Hi. How about if I want to construct the filters with different size of channels? For example, instead of 5 filters all pass 20 channels, i want to do something like[1,10,3,3],[1,20,3,3],[1,30,3,3],[1,30,3,3],[1,10,3,3]. Do you think is possible?

Besides, I wish my question might get your concern. Very appreciated  
[https://discuss.pytorch.org/t/how-to-train-the-paralleled-layers/81317](https://discuss.pytorch.org/t/how-to-train-the-paralleled-layers/81317)

---

_[View the full topic](https://discuss.pytorch.org/t/question-about-group-convolution/17042)._