autograd

Gradient calculator

Automatically computes derivatives of Python and NumPy code for optimization tasks

Efficiently computes derivatives of NumPy code.

GitHub

7k stars
212 watching
917 forks
Language: Python
last commit: 1 day ago
Linked from 1 awesome list


Backlinks from these awesome lists:

Related projects:

Repository Description Stars
auto-differentiation/xad-py Provides automatic differentiation capabilities for numerical computations in Python 12
numpy/numpy A comprehensive library providing efficient numerical computation and data manipulation capabilities for Python-based scientific computing. 28,350
jax-ml/jax A library that provides high-performance numerical computing and machine learning capabilities. 30,744
zou-group/textgrad An autograd engine for textual gradients using large language models to backpropagate gradients. 1,912
facebookresearch/nevergrad A Python toolbox for performing optimization of complex functions without explicit gradient calculation. 3,980
ddbourgin/numpy-ml A collection of machine learning algorithms implemented in NumPy for rapid experimentation and prototyping. 15,789
eriklindernoren/ml-from-scratch Provides implementations of fundamental machine learning models and algorithms from scratch in Python 24,092
pkmr06/pytorch-smoothgrad PyTorch implementation of a technique to improve the interpretability of deep learning models by adding noise to the gradients 168
auto-differentiation/xad A high-performance tool for computing derivatives of complex functions used in various scientific and engineering applications. 337
pbrod/numdifftools A Python library for automatic numerical differentiation of scalar and vector-valued functions. 258
sympy/sympy A computer algebra system written in Python to solve mathematical equations and expressions 13,130
thealgorithms/python A collection of algorithm implementations in Python 195,521
keon/algorithms A collection of Python implementations of various algorithms and data structures. 24,129
kernc/backtesting.py Backtesting tool for evaluating trading strategies 5,650
pydata/bottleneck Provides high-performance NumPy array functions written in C. 1,077