---
id: quantile-forest
version: "1.4.2"
license: Apache-2.0
license_treatment: permissive
maintenance: active
---

# quantile-forest — Quantile regression forests compatible with scikit-learn.

License: permissive · Maintenance: active · Downloads: 78.4K/mo

## What it is and what it does

Quantile-forest is a Python package that implements quantile regression forests (QRF), a non-parametric tree-based ensemble method for estimating conditional quantiles. It extends forest estimators with the ability to predict arbitrary quantiles at inference time without retraining, making it useful for uncertainty quantification and prediction interval estimation on high-dimensional data.

The package provides Cython-optimized estimators that are compatible with and can serve as drop-in replacements for forest regressors. It supports out-of-bag estimation, quantile rank calculation, and proximity counting. Prebuilt wheels are available for Python 3.9–3.14 on macOS, Windows, and Linux, with dependencies on numpy, scipy, and scikit-learn.

Use it for:

- Estimate prediction intervals for regression tasks to quantify uncertainty around point predictions.
- Build uncertainty-aware models for high-dimensional data where conditional quantile estimates are needed.
- Add quantile prediction to existing pipelines without retraining or major refactoring.
- Perform out-of-bag quantile estimation for model validation and robustness assessment.
- Compute quantile ranks and proximity counts for ensemble diagnostics and model interpretability.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Quantile regression forests for estimating conditional quantiles and prediction intervals, with scikit-learn compatibility and Cython optimization.

Yes. The package is actively maintained, has no known vulnerabilities, uses a permissive Apache-2.0 license, and offers a compatible API for quantile regression. Medium install friction is offset by prebuilt wheels and straightforward dependencies on numpy, scipy, and scikit-learn. Suitable for production use in uncertainty quantification and prediction interval tasks.

## Install

pip install quantile-forest
uv add quantile-forest
poetry add quantile-forest

## Installing quantile-forest

Before you install: Medium install friction due to compiled wheels; prebuilt binaries available for Python 3.9–3.14 across macOS, Windows, and Linux, reducing build complexity. Active maintenance with a recent release 54 days ago.

License in practice: Apache-2.0 permissive license allows commercial and private use with minimal restrictions; suitable for most projects.

Quickstart:

pip install quantile-forest

from quantile_forest import RandomForestQuantileRegressor
import numpy
qrf = RandomForestQuantileRegressor()
qrf.fit(X, y)
y_pred = qrf.predict(X, quantiles=[0.025, 0.5, 0.975])

Requires Python 3.9 or later; numpy, scipy, and scikit-learn must be installed.

Verify before relying:

- Performance benchmarks against other quantile regression implementations.
- Scalability limits on dataset size or dimensionality for practical applications.
- Whether out-of-bag estimation and proximity count methods are documented with usage examples.

## Package facts

- License: Apache-2.0 (permissive)
- Python support: supports_current
- Install friction: medium
- Maintenance: active
- Downloads: 78.4K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

quantile regression forests, conditional quantile estimation, prediction intervals uncertainty, tree-based quantile regression, forest extension quantile, non-parametric quantile estimation, high-dimensional uncertainty quantification, quantile-regression, uncertainty-quantification, ensemble-methods

[View on SkillFed](https://skillfed.io/packages/quantile-forest) · [View on PyPI](https://pypi.org/project/quantile-forest/)