# \[Course 4\] 1x1 conv

**URL:** <https://community.deeplearning.ai/t/course-4-1x1-conv/179419>\
**Category:** Convolutional Neural Networks\
**Tags:** coursera-platform\
**Created:** [August 6, 2022, 3:42pm UTC](https://community.deeplearning.ai/t/course-4-1x1-conv/179419 "2022-08-06T15:42:34Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Hagar\_Abouroumia](https://avatars.discourse-cdn.com/v4/letter/h/ec9cab/32.png) [@Hagar\_Abouroumia](https://community.deeplearning.ai/u/Hagar_Abouroumia)\
**Post date:** [August 6, 2022, 3:42pm UTC](https://community.deeplearning.ai/t/course-4-1x1-conv/179419/1 "2022-08-06T15:42:34Z")

</div>

Using 1x1 conv with the same number of channels of the input image act like a non linearly and will help in learning the non linearity operator.  
I do not get how 1x1 conv will help in learning the non linearity operator.  
Thanks

---

<div class="post-metadata">

**Author:** ![balaji.ambresh](https://sea1.discourse-cdn.com/dlai/user_avatar/community.deeplearning.ai/balaji.ambresh/32/128697_2.png) [@balaji.ambresh](https://community.deeplearning.ai/u/balaji.ambresh)\
**Post date:** [August 6, 2022, 4:25pm UTC](https://community.deeplearning.ai/t/course-4-1x1-conv/179419/2 "2022-08-06T16:25:24Z")

</div>

Non-linearity depends on the output activation function. If you see [here](https://www.tensorflow.org/api_docs/python/tf/keras/layers/Conv2D), `activation=None` which means that the output activation function is linear. When you set it to a non-linear function like `relu`, non-linearity comes into play.

Output of a convolution layer has number of channels equal to the number of conv filters used.  
A 1x1 convolution helps changing the number of channels of the output while maintaining the same height and width of the input.