Cannot download workspace

Course Q&A Deep Learning Specialization Improving Deep Neural Networks: Hyperparameter tun
1

image
image1364Ă—791 7.63 KB

how can I download my workspace when the terminal is like that?

2

Hi @Franz90, can you delete all the tar files you have created and try again ?

Let me know if it doesn’t work still.

Thanks,
Mubsi

3

Hi @Mubsi, you mean to delete all the tar files I have created for all the previous assignements? I dont know why but I couldn’t see all the assignements just in one single folder (or at least in #week/#course folders), but I have to open and create a file to download each time.

4

Hi @Franz90, yes, mostly, each assignment is in its own separate workspace. You’d have to have them downloaded individually.

5

@Mubsi mmm I had the same problem even before starting to create those files, but it is a problem related only to 3 of the assignements of the CNN part of the course.

6

Are you following the instructions given in the Coursera article about downloading workspaces ?

7

Yes, but when I go on the jupyter notebook and I use the terminal to download the folders, it shows a different kind of terminal where the commands I use are not runnable

8

I believe those commands work if everything you wan is in a single folder. It then turns that folder into a tar file and you can have it downloaded.

But if everything is not in a single folder, but separate files and folders, then those commands will not really work.

9

Which is the best way to download the entire folder then?

10

It depends on the names of the folders in the workspace. The Coursera article assumes everything is under ~/work, but that is not the way the workspaces are set up here.

Have you tried doing a few “ls” and “pwd” commands to explore? Here’s a linux “find” command that will help you figure out where the assignment is within the file system hierarchy:

find / -name "*.ipynb"

11

You might want to look under /tf for the assignment files. The other thing to note is that symlinks are used in the setup here, so you really want the “h” option on tar in order to “chase” symlinks, but the Coursera article does not include that.

Here’s another recent thread on a similar topic which includes links to yet other threads with information about how to do this.

1 Like