#neural-network

Laplace approximations for Deep Learning.

SciRS2 - Scientific Computing and AI in Rust., providing SciPy-compatible APIs while leveraging Rust's performance, safety, and concurrency features. Unlike traditional scientific libraries

A Data Streaming Library for Efficient Neural Network Training

High-performance quantitative finance in Rust — 120+ stochastic processes, option pricing, calibration, fixed income, risk & copulas, with SIMD/GPU acceleration and Python bindings.

Rust bindings for the C++ api of PyTorch.

OxiZ is a high-performance Satisfiability Modulo Theories (SMT) solver written entirely in Rust. This project is part of an initiative to reimplement Z3 in Pure Rust. Pure Rust is a fundamental requirement - no C/C++ dependencies, no FFI bindings, just clean, safe Rust code.

Pure managed C# tensor library for .NET: hand-written AVX2/AVX-512 SIMD kernels, autodiff and optional CUDA, HIP, OpenCL, Metal, Vulkan and WebGPU backends. Measured benchmarks against TorchSharp, ML.NET, TensorFlow.NET, NumSharp and MathNet are in the README.

Python package for AutoML on Tabular Data with Feature Engineering, Hyper-Parameters Tuning, Explanations and Automatic Documentation

swarmdo — self-contained AI agent orchestration for Claude Code (swarmdo.com)

Deep Learning framework in Go with Tensors, AutoGrad, and GPU acceleration

💥 Fast matrix-multiplication as a self-contained Python library – no system dependencies!