Minitorch: Implement the Torch API from Scratch, from Cornell Tech

#1

Post by rg111 » Wed, Jan 05, 2022, 2:05 PM UTC

Minitorch: Implement the Torch API from Scratch, from Cornell Tech
minitorch.github.io

Re: Minitorch: Implement the Torch API from Scratch, from Cornell Tech

#2

Post by darepublic » Wed, Jan 05, 2022, 4:52 PM UTC

Amazing, I was looking to do something very similar for myself-- dig into pytorch library to see how everything was working.

Re: Minitorch: Implement the Torch API from Scratch, from Cornell Tech

#3

Post by rg111 » Wed, Jan 05, 2022, 5:27 PM UTC

darepublic wrote: post #2

Amazing, I was looking to do something very similar for myself-- dig into pytorch library to see how everything was working.

There is also the book Deep Learning from Scratch.

Re: Minitorch: Implement the Torch API from Scratch, from Cornell Tech

#4

Post by hex9893628af » Wed, Jan 05, 2022, 6:55 PM UTC

rg111 wrote: post #3

darepublic wrote: post #2

Amazing, I was looking to do something very similar for myself-- dig into pytorch library to see how everything was working.

There is also the book Deep Learning from Scratch .

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.

Re: Minitorch: Implement the Torch API from Scratch, from Cornell Tech

#5

Post by kbutler » Wed, Jan 05, 2022, 7:09 PM UTC

hex9893628af wrote: post #4

rg111 wrote: post #3

Earlier quoted context omitted.

There is also the book Deep Learning from Scratch .

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.

If you're doing machine learning, python is a very viable language to actually use.

Re: Minitorch: Implement the Torch API from Scratch, from Cornell Tech

#6

Post by hex9893628af » Wed, Jan 05, 2022, 7:30 PM UTC

kbutler wrote: post #5

hex9893628af wrote: post #4

Earlier quoted context omitted.

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.

If you're doing machine learning, python is a very viable language to actually use.

Not to me it isn’t.

Re: Minitorch: Implement the Torch API from Scratch, from Cornell Tech

#7

Post by asnt » Wed, Jan 05, 2022, 7:47 PM UTC

hex9893628af wrote: post #4

rg111 wrote: post #3

Earlier quoted context omitted.

There is also the book Deep Learning from Scratch .

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.

Maybe not as comprehensive or approachable:

Julia: https://juliadiff.org/ChainRulesCore.jl/stable/FAQ.html#Wher...? (bottom)

OCaml: https://ocaml.xyz/book/

Re: Minitorch: Implement the Torch API from Scratch, from Cornell Tech

#8

Post by cbm-vic-20 » Wed, Jan 05, 2022, 7:50 PM UTC

hex9893628af wrote: post #6

kbutler wrote: post #5

Earlier quoted context omitted.

If you're doing machine learning, python is a very viable language to actually use.

Not to me it isn’t.

No! You must respect the Python ML monoculture! Now hand me my set of MNIST digit images!

Re: Minitorch: Implement the Torch API from Scratch, from Cornell Tech

#9

Post by p1esk » Wed, Jan 05, 2022, 8:10 PM UTC

“from scratch” == “using numpy and numba”

Re: Minitorch: Implement the Torch API from Scratch, from Cornell Tech

#10

Post by cinntaile » Wed, Jan 05, 2022, 9:05 PM UTC

hex9893628af wrote: post #6

kbutler wrote: post #5

Earlier quoted context omitted.

If you're doing machine learning, python is a very viable language to actually use.

Not to me it isn’t.

Almost all neural network frameworks are Python based, so you're severely limiting yourself here.