Hello everyone, good morning!
I’m trying to get in touch with the support (via chat) of the Coursera platform, because when trying to download files from the jupyter notebook from “Week 3 - Programming Task / Image Segmentation with U-Net” from the deep learning specialization, the terminal is also broken, as shown below.
Can anyone tell me if this problem also occurs with other people or would it be just me?
Remembering that I’m following this tutorial to download the files and develop the notebook on my machine and then feed the platform evaluation.
https://medium.com/@kalpeshnpatil/coursera-downloading-jupyter-workspace-files-10090191ca01
That article on Medium is just a copy of the Coursera Help Center article about downloading. Both of them are “generic” and don’t work for the Deep Learning courses. You need to add the “h” option on tar, you need to split and reassemble the tar file and the pathnames they show are not relevant to all DLS assignments. The files are not under ~/work in lots of the assignments. The overall point is you can’t just blindly follow instructions: you have to exercise some judgement. The other issue is that the assignment environment is linux, so you need to have some familiarity with using the linux shell. E.g. here’s a linux command that will help you find the assignment files:
find / -name "*.ipynb"
1 Like
Many thanks again to my great tutor @paulinpaloalto. In fact my knowledge is quite limited regarding linux environments. Until the activity prior to this one, I had been able to perform this generic process normally. In this last activity I couldn’t, and from what I could see, the next ones also have the same problem.
For some reason, they locate the assignment directory (folder) using a different pathname in some of the assignments. Another way to figure out the path it is to insert a new cell in the notebook (Insert → Cell Above) with this command:
!pwd
And then execute that. The exclamation point (!) as the first character in a Jupyter cell says that what follows is a linux shell command, not python. Here’s what you get in the U-Net Notebook:
/tf/W3A2
If you are using the “File → Open”, “New → Terminal” method to get a shell window, then you need to do:
cd /tf/W3A2
Here’s another thread that gives the "tar"and “split” command sequence that you need. You’ll have to adjust the pathnames depending on the assignment.
