Go back     Home Solutions FD-Shifts: A Benchmark for Failure Detection under Distribution Shifts in Image Classification

FD-Shifts: A Benchmark for Failure Detection under Distribution Shifts in Image Classification (fd_shifts)

Authors: Paul F. Jäger, Carsten T. Lüth, Lukas Klein, Till J. Bungert
https://github.com/IML-DKFZ/fd-shifts

This repository contains a framework for benchmarking failure detection methods under distribution shifts in image classification featuring implementations of baseline methods, data sets and metrics. Additionally to a wide range of built-in methods, this framework can be used to evaluate and compare new methods, models and datasets.

FD-Shifts: A Benchmark for Failure Detection under Distribution Shifts in Image Classification Image
Detecting failures should be seen in the context of the overarching goal of preventing silent failures of a classifier by means of confidence scoring functions.
©Paul F. Jäger
License
Apache 2.0

FD-Shifts: A Benchmark for Failure Detection under Distribution Shifts in Image Classification in Helmholtz Imaging CONNECT:

Application
Anomaly Detection

In data analysis, anomaly detection (also referred to as outlier detection and sometimes as novelty detection) is generally understood to be the identification of rare items, events or observations which deviate significantly from the majority …

In Helmholtz Imaging CONNECT
➤Facilities: 0 ➤Centers: 3 ➤Instruments: 0 ➤Labs: 0 ➤Modalities: 0 ➤Experts: 5 ➤Solutions: 1
No facility found.
Center
Deutsches Krebsforschungszentrum (DKFZ)

More than 450,000 people are diagnosed with cancer each year in Germany. Cancer is a disease that poses enormous challenges to research, because every cancer is different and its course can vary immensely even from …

In Helmholtz Imaging CONNECT
➤Applications: 25 ➤Facilities: 0 ➤Instruments: 6 ➤Labs: 11 ➤Modalities: 24 ➤Experts: 21 ➤Solutions: 9
No instrument found.
No modality found.
Lab
Interactive Machine Learning (IML)

Considering human interaction when designing machine learning (ML) systems bears great potential: On the one hand, decision-making in ML systems remains imperfect in practice, thus requiring human interaction for safety-critical applications such as clinical diagnostics. …

In Helmholtz Imaging CONNECT
➤Applications: 0 ➤Facilities: 0 ➤Centers: 2 ➤Instruments: 0 ➤Modalities: 0 ➤Experts: 4 ➤Solutions: 2

Helmholtz Imaging spinning wheel

Please wait, your data is processed