Modularized Implementation of Deep RL Algorithms in PyTorch

Contains high quality implementations of Deep Reinforcement Learning algorithms written in PyTorch

Conformal classifiers, regressors and predictive systems

A library for ready-made reinforcement learning agents and reusable components for neat prototyping

Conformalized Quantile Regression

Quantile Regression Forests compatible with scikit-learn.

Official Implementation for the "Conffusion: Confidence Intervals for Diffusion Models" paper.

👖 Conformal Tights adds conformal prediction of coherent quantiles and intervals to any scikit-learn regressor or Darts forecaster

Valid and adaptive prediction intervals for probabilistic time series forecasting

Support vector machines (SVMs) and related kernel-based learning algorithms are a well-known class of machine learning algorithms, for non-parametric classification and regression. liquidSVM is an implementation of SVMs whose key features are: fully integrated hyper-parameter selection, extreme speed on both small and large data sets, full flexi…

Bringing back uncertainty to machine learning.

R package for Bayesian meta-analysis models, using Stan

Image-to-image regression with uncertainty quantification in PyTorch. Take any dataset and train a model to regress images to images with rigorous, distribution-free uncertainty quantification.

A Julia package for robust regressions using M-estimators and quantile regressions

Lasso Quantile Regression

Our implementation of the paper "A Multi-Horizon Quantile Recurrent Forecaster"

Using an integrated pinball-loss objective function in various recurrent based deep learning architectures made with keras to simultaneously produce probabilistic forecasts for UK wind, solar, demand and price forecasts.

R code for CAViaR model

Slides and notebooks for my tutorial at PyData London 2018

PyTorch - Implicit Quantile Networks - Quantile Regression - C51