Software

Current R packages/python codes and team initiatives associated to the project include:

  • AdaptiveConformal (R package) conformal prediction for time series – paper (author Herb Sussman)
  • declearn A Python framework for federated learning, enabling model training on decentralized data without sharing raw samples.
  • FactoMineR  A widely used R package (4 500 download/day; > 8 million in total, paper  cited > 10000 times) for multivariate data analysis and visualisation, providing a comprehensive set of tools for PCA, and dimensionality reduction methods for mixed data: Correspondence Analysis for contingency tables, Multiple Correspondence Analysis (MCA) for categorical data, Multiple Factor Analaysis (MFA) for multi-blocks/multi-view data etc. FactoMineR is primarily created and maintained by François Husson (Institut Agro Rennes Angers), with contributions from J. Josse
  • metric-learn  A Python package for metric learning, providing efficient implementations of popular algorithms.
  • missMDA  An R package (> 600 000 download) for handling missing values in multivariate analysis. It performs principal component methods (PCA, MCA, MFA) with missing values and it (multiple) imputes continuous, categorical and mixed data.
  • misaem An R package for linear and logistic regression with missing values.
  • misaem A python library for logistic regression with missing values (author: Christophe Muller).

Platforms for open science:  technological developments aimed at structuring and disseminating knowledge and resources for the scientific community. 

Previous packages developed:

  • lori (R package) Imputation of High-Dimensional Count Data using Side Information (author: Geneviève Robin)
  • mimi (R package)  Main Effects and Interactions in Mixed and Incomplete Data (author: Geneviève Robin)
  • SGD-NA (Python code): Debiasing Stochastic Gradient Descent with Missing Completely At Random data (author: Aude Sportisse)

Health projects:

Application for bed allocation monitoring: ICUBAM

ICUBAM provides real-time monitoring of intensive care unit (ICU) bed availability in French hospitals. Data is directly obtained from doctors working inside ICU by sending them SMS with a HTTP link to a form that they can fill in 15 seconds.
The project was co-built by ICU Doctors from CHRU Nancy/Université de Loraine and engineers from INRIA & Polytechnique. It was fleshed out live during the Covid crisis in Eastern France to answer an urgent need for finding available ICU beds in a saturated and deteriorating situation.