rust-autograd - GitLite

KO | EN

gitlite — search

#python #ai #typescript #llm #golang #android #machine-learning #rust #macos #pytorch #mcp #ios #openai #open-source

rust-autograd

(★ 503)

Tensors and differentiable operations (like TensorFlow) in Rust

파일 탐색기

최종 버전 다운로드 (.zip)

- - rust.yml
- cnn_mnist.rs
- download_mnist.sh
- lstm_lm.rs
- mlp_mnist.rs
- mnist_data.rs
- sine.rs
- - adagrad.rs
  - adam.rs
  - mod.rs
  - momentum_sgd.rs
  - sgd.rs
- - activation_ops.rs
  - array_ops.rs
  - basic_source_ops.rs
  - binary_ops.rs
  - blas_ffi.rs
  - const_gen_ops.rs
  - dot_ops.rs
  - gradient_ops.rs
  - graph_ops.rs
  - higher_order_ops.rs
  - hook_ops.rs
  - math_ops.rs
  - mod.rs
  - random_ops.rs
  - reduction_ops.rs
  - xent_ops.rs
- evaluation.rs
- gradient.rs
- graph.rs
- hooks.rs
- lib.rs
- ndarray_ext.rs
- op.rs
- prelude.rs
- tensor.rs
- test_helper.rs
- variable.rs
- lib.rs
- test_array_gen.rs
- test_binary_ops_eval.rs
- test_binary_ops_grad.rs
- test_core.rs
- test_optimizers.rs
- test_tensor_ops_eval.rs
- test_tensor_ops_grad.rs
.gitignore
Cargo.toml
CHANGELOG.md
LICENSE
README.md

🚀 설치 가이드

1. 코드 내려받기

git clone https://github.com/raskr/rust-autograd

깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.

cd rust-autograd

방금 내려받은 프로젝트 폴더 안으로 이동합니다.

2. Rust

보통 추천

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Rust (rustup) rustup으로 설치하면 cargo도 함께 설치됩니다.

cargo build --release

Rust 프로젝트를 컴파일합니다.

cargo run

빌드 후 바로 실행까지 진행합니다.

✅ cargo build가 에러 없이 끝나면 성공입니다. target/ 폴더에 실행 파일이 생성됩니다.

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# autograd

[![build](https://github.com/raskr/rust-autograd/actions/workflows/rust.yml/badge.svg)](https://github.com/raskr/rust-autograd/actions/workflows/rust.yml)
[![Crates.io version](https://img.shields.io/crates/v/autograd.svg)](https://crates.io/crates/autograd)
[![docs.rs](https://docs.rs/autograd/badge.svg)](https://docs.rs/autograd/)

Tensors and differentiable operations backed by [ndarray](https://github.com/rust-ndarray/ndarray).

## Cargo.toml
If you use basic linalg operations, especially matrix multiplications, `blas` feature would be important to speed them up. 
``` toml
[dependencies]
autograd = {"<version>", features = ["blas", "<blas-implementation-choice>"] }
```

`<blas-implementation-choice>` must be one of the following (See also [blas-src](https://github.com/blas-lapack-rs/blas-src))
- `accelerate` macOS only
- `intel-mkl` Intel/AMD CPU only. Includes Vector Mathematics (VM) ops
- `openblas`

## Features
### Reverse-mode automatic differentiation
Here we are just computing partial derivatives of `z = 2x^2 + 3y + 1`.
 ```rust
use autograd as ag;
use ag::tensor_ops::*;

ag::run(|ctx: &mut ag::Context<_>| {
    let x = ctx.placeholder("x", &[]);
    let y = ctx.placeholder("y", &[]);
    let z = 2.*x*x + 3.*y + 1.;

// dz/dy
    let gy = &grad(&[z], &[y])[0];
    println!("{:?}", gy.eval(ctx));   // => Ok(3.)

// dz/dx (requires to fill the placeholder `x`)
    let gx = &grad(&[z], &[x])[0];
    let feed = ag::ndarray::arr0(2.);
    println!("{:?}", ctx.evaluator().push(gx).feed(x, feed.view()).run()[0]);  // => Ok(8.)

// ddz/dx (differentiates `z` again)
    let ggx = &grad(&[gx], &[x])[0];
    println!("{:?}", ggx.eval(ctx));  // => Ok(4.)
});
 ```

### Neural networks
 This crate has various low-level features inspired by tensorflow/theano to train neural networks.
 Since computation graphs require only bare minimum of heap allocations, the overhead is small, even for complex networks.
 ```rust
// MNIST digits classification with multi-layer-perceptron
use autograd as ag;
use ag::optimizers::adam::Adam;
use ag::tensor_ops::*;
use ag::prelude::*;

let mut env = ag::VariableEnvironment::new();

let rng = ag::ndarray_ext::ArrayRng::<f32>::default();

// Register variables in this env.
env.name("w").set(rng.glorot_uniform(&[28 * 28, 10]));
env.name("b").set(ag::ndarray_ext::zeros(&[1, 10]));

let adam = Adam::default("my_adam", env.default_namespace().current_var_ids(), &mut env);

for epoch in 0..3 {  // 0.11 sec/epoch on 2.7GHz Intel Core i5
    env.run(|ctx| {
        let x = ctx.placeholder("x", &[-1, 28*28]);
        let y = ctx.placeholder("y", &[-1]);
        let w = ctx.variable("w");
        let b = ctx.variable("b");
        let z = matmul(x, w) + b;
        let mean_loss = reduce_mean(sparse_softmax_cross_entropy(z, &y), &[0], false);
        let grads = &grad(&[mean_loss], &[w, b]);

// let mut feeder = ag::Feeder::new();
        // feeder.push(x, x_batch).push(y, y_batch);
        // adam.update(&[w, b], grads, ctx, feeder);
    });
}
```

### Abstractions
```rust
use autograd as ag;
use ag::tensor_ops::*;
use ag::ndarray;

// `Tensor::map()`
ag::run(|ctx| {
    let x = ones(&[2, 3], ctx);
    // apply ndarray's methods
    let y = x.map(|x| x.fold_axis(ndarray::Axis(0), 0.0, |acc, x| acc + x));
    let z = x.map(|x| ag::ndarray_ext::zeros(x.shape()));
});

// Hooks
ag::run(|ctx| {
    let x: ag::Tensor<f32> = ones(&[2, 3], ctx).show_shape();
    let y: ag::Tensor<f32> = ones(&[2, 3], ctx).raw_hook(|x| println!("{}", x));
});
```

For detailed, see [documentation](https://docs.rs/autograd/) or
[examples](https://github.com/raskr/rust-autograd/tree/master/examples)