maybe creating a simple "pytorch like library" and training models using that? No?

maybe we can create one ourselves. Posted this a few days ago here. A decent read: https://zekcrates.quarto.pub/deep-learning-library/

Yes i've read your blogposts way back then. Nice work with the gpt-2!!

thanks!!

If you are asking about the "micrograd" video then yes a little bit. "micrograd" is for scalars and we use tensors in the book. If you are reading the book I would recommend to first complete the series or atleast the "micrograd" video.

Its nice! Yeah a lot of the heavy lifting is done by Numpy.

I agree! What NumPy is doing is actually quite beautiful. I was thinking of writing a custom c++ backend for this thing. Lets see what happens this year.

If someone is interested in low level tensor implementation details they could benefit from a course/book “let’s build numpy in C”. No need to complicate DL library design discussion with that stuff.

Yes!!