Hi, I completed the DLS and am wondering if I will still be able to revisit the assignments in the future. Will they be locked out after my last month of paid subscription ends? Is it possible to download all the required materials to run the code natively?
There are topics on the DLS FAQ Thread that cover both of those questions. Please have a look and let us know if any followup questions.
Well, maybe I overstated that a bit: the FAQs cover what happens when you stop paying and how to download everything, but do not cover how to run locally. That turns out to be pretty complicated and there are no official instructions. There are two levels of issues:
- There is quite a bit of software you need to get installed in order to run locally.
- These courses have not been updated since April 2021 and things mutate quite rapidly in the world of python packages in general. That means you will very likely run into “versionitis” problems if you just install the latest version of everything and then try to run.
Here’s a thread which gives some guidance about how to start down the path of addressing those issues, but the full scope of that is up to you. Here’s another useful thread. Please note that this is beyond the scope of the course and mentors are volunteers here. We don’t get paid to answer your questions, so I hope you will understand that asking us to be your personal IT support staff for running things in whatever environment you happen to have is optional. You may get lucky and find community members who like to discuss that sort of thing (as shown in the threads that I linked), but you can’t demand or expect that kind of support from the course staff or the mentors. It doesn’t hurt to ask, but there are no guarantees of useful responses. Just sayin’ … 
Right, that was my concern: running the assignments locally seems to be a major project. It seems as though my best option to revisit any of the assignments is to just renew my subscription. My last month just ended yesterday and now everything is locked. I really wish Coursera would let us keep access to the assignments for a month or two after completing the courses. That was a ton of information to absorb, and I know I won’t retain it without putting it to use.
I’ve already signed up for another Coursera program, so paying for multiple courses at once is just unaffordable. I would be happy to pay for Coursera Plus, but unfortunately, that doesn’t include any of the DeepLearning.AI programs 
I am just a fellow student, so I can’t help with the policy decisions made by Coursera and DLAI.
But it’s definitely a good general point that if you are in this for the long term, it’s worth investing some effort to develop your own method for running experiments and doing work in this space. There are lots of alternatives. You can just use a Python GUI like PyCharm and install the various packages required on your local machine. You can also run Jupyter notebooks locally or on platforms like Google Colab. You could try starting with Colab and just invest the time upfront to learn what you need to do to adapt the notebooks to working there. One advantage of Colab (or even AWS) is that they’ve got scalable hardware power that you can access (for a fee in most cases). If you’re going to want to run training on real datasets, it may not be so easy to do that locally.
1 Like