Librairies ANITI

Depuis sa création, ANITI organise des focus techniques à destination de ses partenaires industriels. Retrouvez la liste des librairies produites par ANITI et accessibles en open source.

Explainability

• XPLIQUE is a Neural Networks Explainability toolbox, composed of various modules implementing various methods such as Attribution Methods, Feature Vizualization, Concepts methods (replay – slides – github). The 2024 version “XPLIQUE UNLEASHED” incorporates new features : new operators, improved attribution methods, automatic concepts extraction and visualization methods(replay – slides – github – tutorials)

• GEMS.AI is a A Python package for AI fairness and interpretability. The approach is to build counterfactual distributions that permit answering “what if?” scenarios. The key principle is that we stress one or more variables of a test set and we then observe how the trained machine learning model reacts to the stress (website – replay – slides  – github)

Robustness

• DEELLIP : Build 1-Lipschitz Neural Networks to guarantee robustness (replay – slides – github)

• DECOMON : a library dedicated to verifying the local stability of developed neural networks. Using advanced techniques such as Linear Relaxation-based Perturbation Analysis (LiRPA), Decomon automatically derives upper and lower bounds for the predictions of a neural network with perturbed inputs. These bounds are calculated using linear relaxations. (replay – slides)

Verification and Validation

• PUNCC a library to predict and quantify the uncertainty of the result of a ML model (replay – slides – github)

• NFLUENCAE a library for tracing the influence back to the data points (replay – slides – github)

Operational domain / domaine d’emploi

OODEEL : Oodeel allows the use of state-of-the-art methods from the Out-Of-Distribution (OOD) detection literature for any image classifier in PyTorch and Tensorflow. It focuses on Post-Hoc OOD methods that apply to already trained models and, therefore, do not require further training (replay – slides – github)

Embedded code generation

ACETONE : A tool to generate C code for neural networks, preserving semantics and guaranteeing a worst-case execution time, with a view to avionics certification objectives. (replay – slides – github)

Planning and Optimization

• Toulbar2 : an efficient solver for discrete constraint optimization (replay – slides – github)
  
• Skicit-decide : an AI toolbox for reinforcement Learning, Automated planning and Scheduling (replay – slides – github)
  
• Discrete-Optimization : Discrete Optimization is a python library to ease the definition and re-use of discrete optimization problems and solvers. ( github)

Computer Vision

• SerreLab: Advanced tools for Computer Vision using bio-inspired feedback mechanisms (replay – slides – github)
  
• PREDIFY: Improve your Computer Vision with a brain-inspired framework (replay – slides – github)
  
• SimODD : An evolutionary testing simulation with Out Of Distribution Images (replay – slides – github)

Robotics

• Pinocchio A fast and flexible implementation of Rigid Body Dynamics algorithms and their analytical derivatives github
  
• Happypose Toolbox and trackers for object pose-estimation. Based on the work CosyPose and MegaPose github
  
• Crocoddyl an optimal control library for robot control under contact sequence. Its solver is based on various efficient Differential Dynamic Programming (DDP)-like algorithms github
  
• Open Dynamic Robot Initiative Open-source hardware and software for torque-controlled robots github