stanfordmlgroup/ngboost
A Python library for probabilistic regression using natural gradient boosting, built on scikit-learn.
Not currently ranked — collecting fresh signals.
star history
NGBoost implements Natural Gradient Boosting for probabilistic prediction, outputting full probability distributions rather than point estimates. It is built on scikit-learn with modular support for different proper scoring rules, distributions, and base learners. The library enables uncertainty quantification in regression and classification tasks through its natural gradient descent approach.
Frequently asked
- What is stanfordmlgroup/ngboost?
- A Python library for probabilistic regression using natural gradient boosting, built on scikit-learn.
- Is ngboost open source?
- Yes — stanfordmlgroup/ngboost is open source, released under the Apache-2.0 license.
- What language is ngboost written in?
- stanfordmlgroup/ngboost is primarily written in Jupyter Notebook.
- How popular is ngboost?
- stanfordmlgroup/ngboost has 1.9k stars on GitHub.
- Where can I find ngboost?
- stanfordmlgroup/ngboost is on GitHub at https://github.com/stanfordmlgroup/ngboost.