# Implementing a custom convolution layer

**URL:** <https://discuss.pytorch.org/t/implementing-a-custom-convolution-layer/5659>\
**Category:** Uncategorized\
**Created:** [August 1, 2017, 6:15am UTC](https://discuss.pytorch.org/t/implementing-a-custom-convolution-layer/5659 "2017-08-01T06:15:08Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![aabobakr](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/a/d2c977/32.png) [@aabobakr](https://discuss.pytorch.org/u/aabobakr)\
**Post date:** [August 1, 2017, 6:15am UTC](https://discuss.pytorch.org/t/implementing-a-custom-convolution-layer/5659/1 "2017-08-01T06:15:08Z")

</div>

I am trying to implement a custom conv layer, so I need to process the input using im2col and col2im utilities to extract input patches according to the kernel size, stride and padding parameters. How to access these utilities in Pytorch ?