# C4W1 CNN back propagation **URL:** \ **Category:** Convolutional Neural Networks\ **Tags:** coursera-platform\ **Created:** [October 12, 2021, 9:43pm UTC](https://community.deeplearning.ai/t/c4w1-cnn-back-propagation/49293 "2021-10-12T21:43:00Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![yunxiangqin](https://avatars.discourse-cdn.com/v4/letter/y/e99b99/32.png) [@yunxiangqin](https://community.deeplearning.ai/u/yunxiangqin)\ **Post date:** [October 12, 2021, 9:43pm UTC](https://community.deeplearning.ai/t/c4w1-cnn-back-propagation/49293/1 "2021-10-12T21:43:00Z")
When calculating db by summing up dZ (optional assignment), why don’t we need to divide by number of training examples m? ---
**Author:** ![jonaslalin](https://sea1.discourse-cdn.com/dlai/user_avatar/community.deeplearning.ai/jonaslalin/32/63_2.png) [@jonaslalin](https://community.deeplearning.ai/u/jonaslalin)\ **Post date:** [November 2, 2021, 9:19am UTC](https://community.deeplearning.ai/t/c4w1-cnn-back-propagation/49293/2 "2021-11-02T09:19:56Z")
Probably because the code computes dJ/db and not dL/db as per lecture videos. I like the former notation better because it is a cleaner solution than to introduce dL/db and dL/dW. We have a recent topic where you can read more: > [@Typo in back prop formula (week3 and week 4)](https://community.deeplearning.ai/t/typo-in-back-prop-formula-week3-and-week-4/50614/5): > > OK, I must be making some kind of blunder here. What I proposed is to define all the partial derivatives with respect to J, instead of what you called L, which is an m-dimensional vector, when we are differentiating with respect to Z's. The point is that \frac{\partial L}{\partial Z} is of dimension m\*\dim Z, which messes things up when we use them for calculations for dW and db's. It will also make things a lot more consistent. In this case, \frac{\partial J}{\partial Z^{[L]}}= \frac{1}{m}(A…