MyTorch – Minimalist autograd in 450 lines of Python
1–10 of 20 posts
Re: MyTorch – Minimalist autograd in 450 lines of Python
#2HmcKk
Re: MyTorch – Minimalist autograd in 450 lines of Python
#3Karpathy’s micrograd did it first (and better); start here: https://karpathy.ai/zero-to-hero.html
Re: MyTorch – Minimalist autograd in 450 lines of Python
#4Karpathy’s micrograd did it first (and better); start here: https://karpathy.ai/zero-to-hero.html
Harsh.
Re: MyTorch – Minimalist autograd in 450 lines of Python
#5Karpathy’s micrograd did it first (and better); start here: https://karpathy.ai/zero-to-hero.html
Why is it better
Re: MyTorch – Minimalist autograd in 450 lines of Python
#6Karpathy’s micrograd did it first (and better); start here: https://karpathy.ai/zero-to-hero.html
Imho, we should let people experiment as much as they want. Having more examples is better than less. Still, thanks for the link for the course, this is a top-notch one
Re: MyTorch – Minimalist autograd in 450 lines of Python
#7Re: MyTorch – Minimalist autograd in 450 lines of Python
#8Earlier quoted context omitted.
Why is it better
Because it's an acclaimed, often cited course by a preeminent AI Researcher (and founding member of OAI) rather than four undocumented python files.
Objective measures like branch depth, execution speed, memory use and correctness of the results be damned.
Re: MyTorch – Minimalist autograd in 450 lines of Python
#9Earlier quoted context omitted.
Because it's an acclaimed, often cited course by a preeminent AI Researcher (and founding member of OAI) rather than four undocumented python files.
Objective measures like branch depth, execution speed, memory use and correctness of the results be damned.
Karpathy's implementation is explicitly for teaching purposes. It's meant to be taken in alongside his videos, which are pretty awesome.
Re: MyTorch – Minimalist autograd in 450 lines of Python
#10Earlier quoted context omitted.
Why is it better
Because it's an acclaimed, often cited course by a preeminent AI Researcher (and founding member of OAI) rather than four undocumented python files.
it being acclaimed is a poor measure of success, theres always room for improvement, how about some objective comparisons?