https://github.com/quansight-labs/udiff

Automatic differentiation with uarray/unumpy.
https://github.com/quansight-labs/udiff

Last synced: 5 months ago
JSON representation

Automatic differentiation with uarray/unumpy.

• Host: GitHub
• URL: https://github.com/quansight-labs/udiff
• Owner: Quansight-Labs
• License: bsd-3-clause
• Created: 2019-10-01T17:24:43.000Z (about 6 years ago)
• Default Branch: master
• Last Pushed: 2021-03-07T13:26:55.000Z (over 4 years ago)
• Last Synced: 2025-04-02T08:06:29.725Z (6 months ago)
• Language: Python
• Size: 1.56 MB
• Stars: 16
• Watchers: 6
• Forks: 11
• Open Issues: 2
• Metadata Files:
  • Readme: README.md
  • Contributing: CONTRIBUTING.md
  • License: LICENSE
  • Code of conduct: CODE_OF_CONDUCT.md

Awesome Lists containing this project

README

          
# `udiff` - Automatic differentiation with uarray/unumpy.

[![Join the chat at https://gitter.im/Plures/uarray](https://badges.gitter.im/Plures/uarray.svg)](https://gitter.im/Plures/uarray?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge) ![language](https://img.shields.io/badge/language-python3-orange.svg) ![license](https://img.shields.io/github/license/Quansight-Labs/udiff)

## Quickstart

```python

import uarray as ua

import unumpy as np

import udiff

from unumpy import numpy_backend

with ua.set_backend(udiff.DiffArrayBackend(numpy_backend), coerce=True):

    x1 = np.reshape(np.arange(1, 26), (5, 5))

    x2 = np.reshape(np.arange(1, 26), (5, 5))

    y = np.log(x1) + x1 * x2 - np.sin(x2)

    print(y)

    print(y.to(x1))

    print(y.to(x2))

```

## Contributing

See [`CONTRIBUTING.md`](CONTRIBUTING.md) for more information on how to contribute to `udiff`.