Coding help for C4_W1 1st lab

Course Q&A Deep Learning Specialization Convolutional Neural Networks
1

trouble with ? padding leading to wrong calculation for Z and Z mean 1st Jypyter lab in C4_W1

2

Hi @Lawrence_Kent

What course are you referent to?

A kindly reminder. Donā€™t forget to post your queries on the right place. This is the only way to us mentors keep tracking the issues and allow us to help all learners in a proper way.

Please, make sure that your follow our guidelines

I hope you understand.

Best regards
elirod

3

Course 4 - Week 1, but which specialization?

1 Like
4

Thanks for your response. i am inquiring the coursera deep learning, course 4 on Convolutional Networks and specifically the 1st Jupyter notebook - convolutional networks step by step. I believe I understand overall what is happening, I have successfully padded the array and was able to calculate the first layer channel. Bur I have run into trouble on the summing of all the Z and channels together. In particular I get the wrong Z and Z mean and in the error messages I get the advice to include all the correct slides, but as far as I can tell the correct slides and pads are included in the calculations. I can send the entire error message but first am I in the right place to ask this question?

Lawrence Kent

5

You posted in ā€œGeneral Discussionsā€, but questions should be created in the forum area for that course. Iā€™ll move your thread there.

Iā€™ll have additional feedback later this afternoon.

6

thank you

7

Please post a screen capture image of the entire error message.

8

HI

The below screenshots capture the error message

Thanks for any help you can give.

image
image1227Ɨ363 16.2 KB

image
image1201Ɨ353 20.9 KB

image
image1214Ɨ353 21 KB

9

In conv_forward(), there is a problem with how your code is computing a_slice_prev.

There are a lot of steps involved.

Check your private messages for instructions.

1 Like
10

Hi: I have trying different ā€˜slidesā€™ to shape the desired output w/o success in the deep learning Convolutional network - step by step lab. I still get the error message ā€˜operands cannot be broadcast togetherā€™ in the ā€˜convolutional forward propagulationā€™. Except for the back propagation (which I would still like to try, I have successfully completed all the other exercise
I have saved the notebook and will send it to you, but want to make sure it is a direct message. Since I am new to this forum, how do I make sure I do that? Thanks

12

I am facing the exact same error, did you find a solution?

ValueError: operands could not be broadcast together with shapes (3,2,4) (3,3,4)
13

NVM I found the error: I was wrongly calculating vert_start, vert_end, horiz_start and horiz_end.

14

Thatā€™s good news!