Course 4/Week2/Assignment 1 - Unable to download resnet50.h5 file

Course Q&A Deep Learning Specialization Convolutional Neural Networks
1

Greetings. I Just finished the Programming Assignment 1 of Week2 of Course 4 (CNN). All is well, I have submittetd the assignment too and it has been graded. I was just trying to download the code to my system for further exploring, I could download all but resnet50.h5. It just displays a blank page when I click on download. I have already verified that kernel is not running (and it is obvious that I could download other files). Please help to resolve.

2

A lot of browsers don’t allow you to download binary files or maybe it’s the fact that file is actually pretty large. There are limits on the size of a file you can download, but it varies by your internet connection.

Here’s a thread with some methods to get around both those issues.

3

Hi Paulinpaloato,

Thanks for the tip, I followed the thread above, and was able to download workspaces of other assignments, but not this one. When I try to open a new terminal, it opens, but I see nothing in that folder! The current working directory is /home/work but it appears to be empty. I am not sure if it is due to a permission issue on the folder/files and hence I can’t view them.
Just as fyi, below are the commands which worked on the other folders (W1A1, W1A2 and W2A2)
tar -czf C4W1A1.tar.xz *
split -b 50m C4W1A1.tar.xz C4W1A1.tar.xz.part.

I need help for W2A1. Thanks

I see that a similar issue exists for W3 Assignment 2 and W4 Assignment 2 as well. Not sure if it is just me.

4

In some of the assignments, they have a different directory structure and the assignment files are not under /home. Here’s a linux command that will show you where the notebook files are:

find / -name "*.ipynb"

Then you have to “cd” to the appropriate directory before you do the rest of the commands.

5

Thank you, this helped. I could locate the folder.

For the benefit of other/new learners, just a suggestion, if these little guidelines can be included part of the programming assignment instructions, it can save time both ends, especially that we should lookout for the files in a different folder. As a user I would not know if I even have access to search in the root folders though I was aware of the find command.

As for as I am concerned, this issue is resolved and can be considered as “Closed”.

6

I’m glad to hear that you found the files. But note that there are no “official” instructions for how to do this. If we ever create them, we will definitely mention this point.

Also note that there are other perhaps easier ways to avoid this problem. You can run shell commands from the notebook. Just do add a new cell and then preface any shell commands with an exclamation point (!). These are called “bang escapes”. Just a little nerd humor. :nerd_face: Here’s an example that would have answered the “find” question:

!pwd

Give it a try and watch what happens. Of course you could also just go “all in” on that strategy and run the tar commands and the rest of it from cells in the notebook and then you won’t have to worry about the location of the files in the terminal. The downside of that strategy is that you then have to comment out those commands so that you don’t run them every time. Or put them in a different copy of the notebook than the primary one. Just some additional thoughts …

7

Thank you for this additional information.

I do appreciate it.

Regards,
NAGARAJA
9980162087

8

Inspired/motivated by your guidance above, I created this simple file - which can be included part of the workspace (I have named it Save_your_workspace.ipynb).

The only thing that will change in the script below is the filename. Some one could easily do a find and replace for that course/week/assignment.

#BEGIN
!tar -czf C5W1A3.tar.xz *
file_data = !ls -l C5W1A3.tar.xz 
file_size = int(file_data.s.split()[4])
print(file_data)
if file_size > 50*1024*1024:
    !split -b 50m C5W1A3.tar.xz C5W1A3.tar.part.
    print("As the file size is > 50 MB, the file has been split into parts. Look for file names with *part* and download them individually.")
else:
    print("Your workspace has been saved into C5W1A3.tar.xz. You may download the same.")
#END

I hope I am not violating the coursera honor code here as this is a very generic code (nothing technical). Please feel free to remove if it does violate. I am just trying to add my 2 cents here.

I used this for the first week’s assignments of course 5 :slight_smile:

9

That looks great! Thanks for showing everyone how to do this in a nice way. Since it is structured with the “!” escapes, they can just insert a new cell in any one of their assignment notebooks and “copy/paste” in that code and then change the file names.

It’s perfectly fine to share this kind of code, since it’s not part of the solution to any of the assignments. Sharing answers to the assignments or the quizzes is the only thing that’s off limits.

Thanks again!