Metadata-Version: 2.2
Name: herding
Version: 0.0.2
Summary: Kernel herding
Home-page: https://github.com/microprediction/herding
Author: microprediction
Author-email: peter.cotton@microprediction.com
License: MIT
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.12
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy
Requires-Dist: randomcov
Dynamic: author
Dynamic: author-email
Dynamic: classifier
Dynamic: description
Dynamic: description-content-type
Dynamic: home-page
Dynamic: license
Dynamic: requires-dist
Dynamic: summary

# herding
Kernel herding is a kind of quasi-monte carlo where samples are chosen successively to approximate known kernel moment constraints. 

In the [video](https://github.com/microprediction/herding/tree/main/docs/assets/video) also linked below, the green dots are landmark locations `y` and the expected value of a gaussian kernel `E[ker(x,y)]` is assumed known for the desired distribution for `x`, which happens to be Gaussian.   

See [examples](https://github.com/microprediction/herding/tree/main/examples)


[![Herding Demo](docs/assets/images/herding_video_thumb.png)](docs/assets/video/herding_video_low_res.mp4)


