ngboost and XGBoostLSS
These are **competitors**: both extend gradient boosting frameworks to output full probability distributions rather than point predictions, offering alternative approaches to the same probabilistic prediction problem.
About ngboost
stanfordmlgroup/ngboost
Natural Gradient Boosting for Probabilistic Prediction
This library helps data scientists and machine learning engineers create models that predict not just a single outcome, but a full range of possible outcomes and their likelihoods. You provide structured data with features and a target variable, and it outputs a model that offers a probability distribution for future predictions, rather than just a point estimate. This is useful for anyone needing to understand the uncertainty in their predictions.
About XGBoostLSS
StatMixedML/XGBoostLSS
An extension of XGBoost to probabilistic modelling
This tool helps data scientists and analysts make more robust predictions by forecasting the full range of possible outcomes, not just a single value. It takes in your dataset with various features and outputs not only a prediction, but also the likelihood of different potential results, including prediction intervals and quantiles. This is ideal for professionals who need to understand the uncertainty and risk associated with their forecasts.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work