[HN Gopher] Minitorch: Implement the Torch API from Scratch, fro...
___________________________________________________________________
Minitorch: Implement the Torch API from Scratch, from Cornell Tech
Author : rg111
Score : 67 points
Date : 2022-01-05 14:05 UTC (8 hours ago)
(HTM) web link (minitorch.github.io)
(TXT) w3m dump (minitorch.github.io)
| p1esk wrote:
| "from scratch" == "using numpy and numba"
| darepublic wrote:
| Amazing, I was looking to do something very similar for myself--
| dig into pytorch library to see how everything was working.
| rg111 wrote:
| There is also the book _Deep Learning from Scratch_.
| hex9893628af wrote:
| Are there any resources that use languages other than python?
| I would prefer to work with code examples in a language that
| I might actually use.
| asnt wrote:
| Maybe not as comprehensive or approachable:
|
| Julia: https://juliadiff.org/ChainRulesCore.jl/stable/FAQ.h
| tml#Wher...? (bottom)
|
| OCaml: https://ocaml.xyz/book/
| kbutler wrote:
| If you're doing machine learning, python is a very viable
| language to actually use.
| hex9893628af wrote:
| Not to me it isn't.
| systemsignal wrote:
| What language do you use?
| cinntaile wrote:
| Almost all neural network frameworks are Python based, so
| you're severely limiting yourself here.
| cbm-vic-20 wrote:
| No! You must respect the Python ML monoculture! Now hand
| me my set of MNIST digit images!
___________________________________________________________________
(page generated 2022-01-05 23:02 UTC)