Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Efficiently computes derivatives of NumPy code.
| Date | Stars |
|---|---|
| 2026-07-24 | 7513 |
| 2026-07-25 | 7513 |
| 2026-07-28 | 7513 |
| 2026-07-30 | 7513 |
| 2026-08-06 | 7513 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
15.0
growth rate 0.00%/day
# Autograd [![Checks status][checks-badge]][checks-url] [![Tests status][tests-badge]][tests-url] [![Publish status][publish-badge]][publish-url] [![asv][asv-badge]](#) [publish-badge]: https://github.com/HIPS/autograd/actions/workflows/publish.yml/badge.svg [checks-badge]: https://github.com/HIPS/autograd/actions/workflows/check.yml/badge.svg [tests-badge]: https://github.com/HIPS/autograd/actions/workflows/test.yml/badge.svg [asv-badge]: http://img.shields.io/badge/benchmarked%20by-asv-green.svg?style=flat [publish-url]: https://github.com/HIPS/autograd/actions/workflows/publish.yml [checks-url]: https://github.com/HIPS/autograd/actions/workflows/check.yml [tests-url]: https://github.com/HIPS/autograd/actions/workflows/test.yml Autograd can automatically differentiate native Python and Numpy code. It can handle a large subset of Python's features, including loops, ifs, recursion and closures, and it can even take derivatives of derivatives of derivatives. It supports reverse-mode differentiation (a.k.a. backpropagation), which means it can efficiently take gradients of scalar-valued functions with respect to array-valued arguments, as well as forward-mode differentiation, and the two can be composed arbitrarily. The main intended application of Autograd is gradient-based optimization. For more information, check out the [tutorial](docs/tutorial.md) and the [examples directory](examples/). Example use: ```python >>> import autograd.numpy as np # Thinly-wrapped numpy >>> from autograd import grad # The only autograd function you may ever need >>> >>> def tanh(x): # Define a function ... return (1.0 - np.exp((-2 * x))) / (1.0 + np.exp(-(2 * x))) ... >>> grad_tanh = grad(tanh) # Obtain its gradient function >>> grad_tanh(1.0) # Evaluate the gradient at x = 1.0 np.float64(0.419974341614026) >>> (tanh(1.0001) - tanh(0.9999)) / 0.0002 # Compare to finite differences np.float64(0.41997434264973155) ``` We can continue to differentiate as many times as we like, and use numpy's vectorization of scalar-valued functions across many different input values: ```python >>> from autograd import elementwise_grad as egrad # for functions that vectorize over inputs >>> import matplotlib.pyplot as plt >>> x = np.linspace(-7, 7, 700) >>> plt.plot(x, tanh(x), ... x, egrad(tanh)(x), # first derivative ... x, egrad(egrad(tanh))(x), # second derivative ... x, egrad(egrad(egrad(tanh)))(x), # third derivative ... x, egrad(egrad(egrad(egrad(tanh))))(x),) # fourth derivative >>> plt.show() ``` <img src="examples/tanh.png" width="600"> See the [tanh example file](examples/tanh.py) for the code. ## Documentation You can find a tutorial [here.](docs/tutorial.md) ## End-to-end examples * [Simple neural net](examples/neural_net.py) * [Convolutional neural net](examples/convnet.py) * [Recurrent neural net](examples/rnn.py) * [LSTM](examples/lstm.py) * [Neural Turing Machine](https://github.com/DoctorTeeth/diffmem/blob/512aadeefd6dbafc1bdd253a64b6be192a435dc3/ntm/ntm.py) * [Backpropagating through a fluid simulation](examples/fluidsim/fluidsim.py) <img src="examples/fluidsim/animated.gif" width="400"> * [Variational inference in Bayesian neural network](examples/bayesian_neural_net.py) * [Gaussian process regression](examples/gaussian_process.py) * [Sampyl, a pure Python MCMC package with HMC and NUTS](https://github.com/mcleonard/sampyl) ## How to install Install Autograd using Pip: ```shell pip install autograd ``` Some features require SciPy, which you can install separately or as an optional dependency along with Autograd: ```shell pip install "autograd[scipy]" ``` ## Authors and maintainers Autograd was written by [Dougal Maclaurin](https://dougalmaclaurin.com), [David Duvenaud](https://www.cs.toronto.edu/~duvenaud/), [Matt Johnson](http://people.csail.
Excerpt of 4,780 characters
Read on GitHub
Matthew Johnson · Google · United States
438
Dougal Maclaurin · Google Brain · United States
382
David Duvenaud · University of Toronto
279
Jamie Townsend · Anthropic · United Kingdom
172
85
Agriya Khetarpal · @Quansight · India
27
Gianluca Campanella · @thetradedesk · United Kingdom
20
Yu Kobayashi · Japan
17
10
Fabian Joswig · DeepL
9
Dan Raviv · Sound Radix
8
Kenneth C. Arnold
6
5
5
5
Jasper Snoek
5
4
Matti Picus · Quansight Labs
4
David Chiang · University of Notre Dame
3
3
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:fc81ec5ed38119a9, topic:deep-learning, topic:neural-network, topic:jax