reverse-mode-autodiff

Differentiable matrix factorizations using ImplicitDifferentiation.jl.

Parallel Reverse Mode Automatic Differentiation in C# for Custom Neural Network Development

A modular C++17 framework for automatic differentiation

C++ automatic differentiation library with no dependencies and arbitrary higher order derivatives, stand-alone, header only

Tiny autograd engine written in Go.

Reverse-mode autodiff for Clojure

Reverse-mode automatic differentiation from scratch — a scalar autograd engine and neural network library in 141 lines of dependency-free Python. Gradients verified against PyTorch.

Yet another automatic differentiation engine to perform efficient and analytically precise partial differentiation of mathematical expressions.

Understanding neural network libraries and the automatic gradient computations (autograd) in the backward pass

A scalar reverse-mode automatic differentiation library with neural network support, built from scratch with PyTorch-like API.

nanograd is a minimal deep learning framework implementing reverse-mode automatic differentiation and neural network layers from scratch in Python.

A library for reverse mode automatic differentiation

Reverse-mode autograd and small neural-net framework in pure NumPy. Higher-order gradients and a tiny GPT.

Lightweight Python package for automatic differentiation

A lightweight, zero-dependency reverse-mode automatic differentiation engine that implements the chain rule over computational graphs from scratch.

A C++23 library for automatic differentiation; python bindings available