Lib.rs

› Science › Machine learning › Burn
#deep-learning #automatic-differentiation #machine-learning #data

no-std burn-autodiff

Automatic differentiation backend for the Burn framework

by Nathaniel Simard and 97 contributors

18 releases (breaking)

new 0.16.0 Jan 14, 2025
0.15.0 Oct 28, 2024
0.14.0 Aug 27, 2024
0.13.2 May 3, 2024
0.3.0 Nov 20, 2022

#949 in Machine learning

Download history 1549/week @ 2024-09-27 1340/week @ 2024-10-04 1675/week @ 2024-10-11 1384/week @ 2024-10-18 1879/week @ 2024-10-25 1817/week @ 2024-11-01 1216/week @ 2024-11-08 1195/week @ 2024-11-15 1452/week @ 2024-11-22 4328/week @ 2024-11-29 3366/week @ 2024-12-06 2568/week @ 2024-12-13 3783/week @ 2024-12-20 3031/week @ 2024-12-27 2844/week @ 2025-01-03 2126/week @ 2025-01-10

12,121 downloads per month
Used in 35 crates (8 directly)

MIT/Apache

2MB
41K SLoC

Burn Autodiff

Burn autodiff backend

Current Crates.io Version license

For now only first order reverse mode autodiff is supported.

lib.rs:

Burn Autodiff

This autodiff library is a part of the Burn project. It is a standalone crate that can be used to perform automatic differentiation on tensors. It is designed to be used with the Burn Tensor crate, but it can be used with any tensor library that implements the Backend trait.

Dependencies

~6MB
~103K SLoC