WCCI Performance Prediction Challenge

The challenge is over, but a new challenge is on-going using the same datasets, check it out!

CLOP-models-only-2

Submitted by Juha Reunanen

Like CLOP-models-only-1, with two differences:

1) Validation labels used for all datasets.

2) Model search process much less intensive, which may lead to inferior results.

Dataset Balanced Error Test guess Guess error Test score Area Under Curve
Train Valid Test Train Valid Test
ada 0.193 0.1896 0.2086 0.1962 0.0123 0.2209 0.8915 0.8924 0.8796
gina 0 0 0.098 0.0962 0.0018 0.0992 1 1 0.9631
hiva 0.2989 0.35 0.3287 0.3303 0.0016 0.3291 0.7412 0.6521 0.6884
nova 0.0417 0.034 0.0933 0.0893 0.004 0.0964 0.999 0.9957 0.9818
sylva 0.0092 0.0012 0.0344 0.0302 0.0041 0.0383 0.9998 0.9998 0.9977
Overall 0.1086 0.115 0.1526 0.1485 0.0048 0.1568 0.9263 0.908 0.9021

This entry is a complete valid challenge entry.