Course 4 #Week 01 : #parameters in CONV2

Course Q&A Deep Learning Specialization Convolutional Neural Networks
1

How number of parameters be 2416 to CONV2(f=5,s=1);
It should be 416 ( 5 * 5 * 16), How it be 2416?

Clarifications about Upcoming CNN Example Video | Coursera

2

Don’t forget that the input to that layer has 6 channels. So each filter is 5 x 5 x 6 + 1 for the bias unit. 151 * 16 = 2416.

1 Like
3

Thank You