I have used stride to jump over n_H and n_W and test one gave me all test passed but in the second test it say AssertionError: Wrong value for A[0] in mode max. Make sure you have included stride in your calculation here this output [image]

The key difference in the test cases is the first one has stride = 1 and the second one has stride = 2. The stride works the same way here that it did in conv_forward: you use it to compute vert_start and horiz_start. It is not involved in the ranges of the loops. You got some all zero values, which…

Pool_forward

Course Q&A Deep Learning Specialization Convolutional Neural Networks

saifkhanengr February 26, 2024, 7:39am 2

Maybe this guide helps.

Topic		Replies	Views
Course 4 ,Week 1 Assignement 1 - All test pass but 50/100 Convolutional Neural Networks coursera-platform	9	583	December 16, 2021
Week 1: Conv_forward Convolutional Neural Networks coursera-platform	2	517	August 30, 2022
Course 4 Week 1 Assignment 1 (50/100) Convolutional Neural Networks coursera-platform	3	558	January 9, 2022
Course 4 week1 conv_forward AssertionError: Wrong Z mean. Expected: -0.5384027772160062 got: 0.0 Convolutional Neural Networks coursera-platform	2	623	December 29, 2021
Course 4 week 1 Assignment 1 Convolutional Neural Networks coursera-platform	3	602	December 14, 2021

Pool_forward

Related topics