Lesson 10 (part 2) preview

We are making the full videos of lessons 9 and 10 of the new “From Deep Learning Foundations to Stable Diffusion” course available as a special preview of the new course! Here’s all the resources for lesson 10:

Lesson resources

Links from the lesson


Enjoying L9 & L10 videos.
How have you found such important papers immediately among many papers published?

1 Like

Hi Jeremy, thanks so much for putting out these videos. I wasn’t able to do the most recent online course and I was wondering if you have an idea (even vaguely) of when we might get the remaining Part 2 lesson videos. Thank you!

1 Like

On the registration page for the part 2 course

it says in the FAQ section:
1. Will a recording of the course sessions be made available after?
Registered participants will have immediate access to the recordings. Public access to the course lectures and materials is typically available 2 or 3 months after completion.

So I guess the part 2 videos will be available around end of January or in February.

@jeremy I was waiting for this course but somehow missed the notice when it opened. I’d have gladly paid for being able to attend it live.

Isn’t there any chance to pay for advance access? I’d rather pay to get access to the training now than having to wait to the end of January/February.

Unfortunately not - the university closed registrations a few weeks ago.

Understand. Had to try though ;-). I’ll wait eagerly for the public release then. Thanks a lot for everything you do Jeremy!


@jeremy Any hint about when part 2 will be publicly released? I am taking time off work right now to work on part 1 (and implement a few ideas that I have), and I’m hoping that part 2 will be available not too long after I finish the first part.

I saw that you released a preview of part 2, which is awesome! And that you said in September that it will be out publicly in early 2023.

Obviously, no pressure to give any more info if it’s not something you want to share. But if you are comfortable with it, I’d just love to know if it’ll be in the next couple of months or if it will likely be later than that.

Regardless of whether you can give any additional info, I hugely appreciate the time and effort you’ve put into making these amazing courses :). Thanks so much!


It’ll be within the next couple of months.


Awesome, thanks very much! That’s helpful - I was debating whether to do one of the old versions of part 2, but instead I’ll wait for the new one

1 Like

Thank you for the time spent on these courses, these courses are the best I’ve come across for my purposes, the practice-oriented approach is very helpful in learning the material.
You help make the world a better place!) :grinning:

1 Like

the chunks function you write here is commonly-enough used that something like it is about to be added to the python standard library in the next version. It’s called itertools.batched, and the reference implementation† looks a lot like you were doing with islice:

def batched(iterable, n):
    "Batch data into tuples of length n. The last batch may be shorter."
    # batched('ABCDEFG', 3) --> ABC DEF G
    if n < 1:
        raise ValueError('n must be at least one')
    it = iter(iterable)
    while batch := tuple(islice(it, n)):
        yield batch

Why is it there? Because a few months ago somebody started a discussion on the python forums, and made the case to get it included.

† the actual implementation is in c, but it’s equivalent to that python code

1 Like