burn/crates/burn-tensor
Dilshod Tadjibaev e8b915a2da
Enhance slice operation to support more range variation (#1989)
* Enhance slice operation to support more range variation

* Fix doc clippy

* Fixed doc test

* Fix flipped attribute names

* Fix clippy
2024-07-08 13:34:25 -05:00
..
src Enhance slice operation to support more range variation (#1989) 2024-07-08 13:34:25 -05:00
Cargo.toml Add static tensor quantization (#1963) 2024-07-08 10:16:58 -04:00
LICENSE-APACHE Update licenses symlinks (#1613) 2024-04-12 14:43:58 -04:00
LICENSE-MIT Update licenses symlinks (#1613) 2024-04-12 14:43:58 -04:00
README.md [refactor] Move burn crates to their own crates directory (#1336) 2024-02-20 13:57:55 -05:00
env.bash [refactor] Move burn crates to their own crates directory (#1336) 2024-02-20 13:57:55 -05:00

README.md

Burn Tensor

Burn Tensor Library

Current Crates.io Version license

This library provides multiple tensor implementations hidden behind an easy to use API that supports reverse mode automatic differentiation.

Features

  • Flexible
  • CPU + GPU 🙏
  • Multi-Threads 🚀
  • Intuitive Usage 😌
  • No Global State 🚫
  • Multiple Backends 🦾
  • Reverse Mode Autodiff 🔥

Backends

For now, three backends are implemented, and some more are planned.

Autodiff

Automatic differentiation is implemented as just another tensor backend without any global state. It's possible since we keep track of the order in which each operation as been executed and the tape is only created when calculating the gradients. To do so, each operation creates a new node which has a reference to its parent nodes. Therefore, creating the tape only requires a simple and efficient graph traversal algorithm.

    let x = AutodiffTensor::from_tensor(x_ndarray);
    let y = ADtodiffTensor::from_tensor(y_ndarray);

    let z = x.matmul(&y);

    let grads = z.backward();

    let x_grad = x.grad(&grads);
    let y_grad = y.grad(&grads);

Cuda

To run with CUDA set TORCH_CUDA_VERSION=cu121.

Notes

This crate can be used alone without the entire burn stack and with only selected backends for smaller binaries.

Feature Flags

This crate can be used without the standard library (#![no_std]) with alloc by disabling the default std feature.

  • std - enables the standard library.
  • burn-tensor-testgen - enables test macros for generating tensor tests.