Good Classification Measures and How to Find Them
This repository contains supplementary materials for the paper "Good Classification Measures and How to Find Them" (NeurIPS 2021).
How can I run it?
To run the experiments, you will need numpy
, and scikit-learn
libraries.
Use the input data files, provided in /data folder, and these scripts:
binary_indicies_bruteforce.py [
-- to reproduce the table N3] meteo_inversion_matrix.py [full|10m|2h]
-- to reproduce tables N4, N10, N11multiclass_indicies_imagenet.py
-- to reproduce tables N5 and N12multiclass_indicies_sst.py
-- to reproduce the table N6multiclass_indicies_yeast.py
-- to reproduce tables N13 and N14