Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Roadmap #1

Closed
2 of 5 tasks
Wuelle opened this issue Oct 6, 2021 · 2 comments
Closed
2 of 5 tasks

Roadmap #1

Wuelle opened this issue Oct 6, 2021 · 2 comments

Comments

@Wuelle
Copy link
Owner

Wuelle commented Oct 6, 2021

  • Add a working implementation for basic dual number operations (Add, AddAssign etc)
  • Allow interfacing with regular numbers implementing num_traits::Float
  • Use standard floating point types for weights/biases, these don't require gradients.
    see Any way to dot arrays of different type? rust-ndarray/ndarray#1067
  • Implement SGD using Dual Numbers
  • Test Performance + Memory usage. Dual Numbers might be really bad
@Wuelle Wuelle pinned this issue Oct 6, 2021
@archuser555
Copy link

Can I Help? :-)

@Wuelle
Copy link
Owner Author

Wuelle commented Sep 27, 2022

Hey, I haven't really worked on this project for a while (and it's probably fair to consider it dead ^^ )
I did some rough calculations and my initial idea of using dual numbers for automatic differentiation would lead to a completely ridiculous amount of memory required. I really appreciate the though but if you want to build a NN library you're probably better of starting a new repo. I might pick this up again at some point but I don't have the time to work on this right now. You are welcome to fork this and do whatever with it.

@Wuelle Wuelle closed this as completed May 16, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants