rusty-gradients 0.2.0

A full-stack deep learning framework in Rust for training and deploying Transformer models. Features multi-backend support (CPU/CUDA/Metal/WASM), 62x GPU acceleration, Safetensors serialization, and BPE tokenization.
Documentation 
1
2
3
4
5
6
7
8
9
10

//! Operations V2 - операции с autograd для TensorV2
//!
//! Все операции автоматически поддерживают backward pass.

pub mod basic;
pub mod matmul;

// Re-export основных операций
pub use basic::{add_op, mul_op, sub_op};
pub use matmul::matmul_op;