ngboost
Library for probabilistic predictions via gradient boosting.
Decision gist · record as of 2026-08-14
Yes. NGBoost is actively maintained, has low install friction, carries a permissive Apache 2.0 license, and fills a clear niche: probabilistic prediction with uncertainty quantification. Install it if you need prediction intervals, distribution outputs, or uncertainty estimates alongside your regression predictions.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Low friction: pure Python wheel, active maintenance (last commit 2026-07-01), and supports current Python versions (3.9–3.13).
- Dependencies are all standard scientific stack (scikit-learn, numpy, scipy, matplotlib, tqdm, lifelines, sympy).
License · maintenance · safety
Apache-2.0 (permissive) — Apache License 2.0 is permissive; you can use, modify, and distribute ngboost freely in commercial and private projects with minimal restrictions.
last release 2026-06-26 (49 days) · last repo commit 2026-07-01 · 1,888 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 179,315 downloads/mo, #10,175 on PyPI
Alternatives
Verify before relying
pip install ngboost
from ngboost import NGBRegressor
ngb = NGBRegressor().fit(X_train, Y_train)
Y_preds = ngb.predict(X_test)
Y_dists = ngb.pred_dist(X_test)- Whether NGBoost supports classification tasks in addition to regression.
- Performance characteristics and scalability limits on large datasets.
- Availability and ease of adding custom distributions or scoring rules beyond built-in options.
What it is and what it does
NGBoost is a gradient boosting library that outputs full probability distributions for each prediction rather than single point estimates. Built on scikit-learn, it implements the Natural Gradient Boosting algorithm, allowing you to quantify prediction uncertainty and choose among different distributions and scoring rules to match your problem. The library depends on numpy, scipy, scikit-learn, matplotlib, tqdm, lifelines, and sympy.
You use it like a standard estimator: fit on training data, then call predict() for point estimates or pred_dist() to get the full predictive distribution at each test point. This makes it useful when you need not just a prediction but also a measure of confidence or uncertainty around that prediction, such as in risk-sensitive applications or when building prediction intervals.
Use it for
- Build regression models that output prediction intervals and uncertainty estimates alongside point forecasts.
- Quantify model confidence in high-stakes domains like medical or financial prediction where uncertainty matters.
- Compare different probability distributions and scoring rules for the same dataset to find the best fit.
- Extend NGBoost with custom distributions or scoring rules for domain-specific probabilistic prediction tasks.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes.
NGBoost is actively maintained, has low install friction, carries a permissive Apache 2.0 license, and fills a clear niche: probabilistic prediction with uncertainty quantification. Install it if you need prediction intervals, distribution outputs, or uncertainty estimates alongside your regression predictions.
Install
ngboost on PyPI
Before you install
Low friction: pure Python wheel, active maintenance (last commit 2026-07-01), and supports current Python versions (3.9–3.13). Dependencies are all standard scientific stack (scikit-learn, numpy, scipy, matplotlib, tqdm, lifelines, sympy).
License in practice
Apache License 2.0 is permissive; you can use, modify, and distribute ngboost freely in commercial and private projects with minimal restrictions.
Quickstart
pip install ngboost
from ngboost import NGBRegressor
ngb = NGBRegressor().fit(X_train, Y_train)
Y_preds = ngb.predict(X_test)
Y_dists = ngb.pred_dist(X_test)Verify before relying
- Whether NGBoost supports classification tasks in addition to regression.
- Performance characteristics and scalability limits on large datasets.
- Availability and ease of adding custom distributions or scoring rules beyond built-in options.
Package facts
| License | Apache-2.0 permissive |
| Python support | Supports the current Python release <3.15,>=3.9 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 7 packagesscikit-learnnumpyscipytqdmlifelinessympymatplotlib |
| Maintenance | Actively maintained 49 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 179,315 / month, #10,175 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | License :: OSI Approved :: Apache Software LicenseOperating System :: OS IndependentProgramming Language :: Python :: 3Programming Language :: Python :: 3.10Programming Language :: Python :: 3.11Programming Language :: Python :: 3.12Programming Language :: Python :: 3.13Programming Language :: Python :: 3.9 |
Evidence: ngboost-0.5.11-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “probabilistic gradient boosting”
- ngboostNGBoost implements Natural Gradient Boosting for probabilistic…
- pytabkitPyTabKit provides scikit-learn interfaces to modern tabular machine…
- lightgbmLightGBM is a gradient boosting framework for classification,…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also scikit-multilearn · xgboost-cpu · catboost · tensorflow-probability · lightgbm · xgboost · quantile-forest · MAPIE · forestci · properscoring