Stars
A toolkit to calibrate predictive algorithms to achieve risk control.
A runway dataset and a generator of synthetic aerial images with automatic labeling.
Python library for classifier calibration
Unifying Variational Autoencoder (VAE) implementations in Pytorch (NeurIPS 2022)
Machine Learning algorithms implemented from scratch
code for "Isolating Sources of Disentanglement in Variational Autoencoders".
Model-agnostic posthoc calibration without distributional assumptions
A Python toolbox for conformal prediction research on deep learning models, using PyTorch.
Flexible and scalable template based on PyTorch Lightning + Hydra. Efficient workflow and reproducibility for rapid ML experiments.
Modelling electrical motor dynamics using neural networks.
Competitive DL-based model on the M4 competition dataset
Experiments used in "Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning"
Estimation of the confidence measure for anomaly detectors, as explained in the paper "Quantifying the Confidence of Anomaly Detectors in Their Example-Wise Predictions" (ECML-PKDD 2020).
A repository for some common operations for everyone
Library for Meta-Recognition and Weibull based calibration of SVM data.
Results of the "Ensembles of offline changepoint detection methods" research to reproduce
PyTorch implementation of "Supervised Contrastive Learning" (and SimCLR incidentally)
Implementation of an Openset Recognition algorithm.
Air traffic generation with VAE
PyTorch Lightning + Hydra. A very user-friendly template for ML experimentation. ⚡🔥⚡
N2D2 is an open source CAD framework for Deep Neural Network simulation and full DNN-based applications building.
Uncertainty Toolbox: a Python toolbox for predictive uncertainty quantification, calibration, metrics, and visualization
Code to accompany the paper Radial Bayesian Neural Networks: Beyond Discrete Support In Large-Scale Bayesian Deep Learning
Bayesian Neural Networks to predict RUL on N-CMAPSS