WCCI Performance Prediction Challenge

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

Final #3 (corrected)

Submitted by Gavin Cawley

model selection based on cross-validation BER, estimation based on cross-validation (likely to be a little biased biased).

ADA #135
GINA #123
HIVA #044
NOVA #041
SYLVA #087

Dataset Balanced Error Test guess Guess error Test score Area Under Curve
Train Valid Test Train Valid Test
ada 0.1501 0.1494 0.185 0.1737 0.0113 0.1963 0.9248 0.926 0.8965
gina 0 0 0.0461 0.047 0.0009 0.0466 1 1 0.99
hiva 0.019 0.0257 0.2757 0.2692 0.0065 0.2797 0.9993 0.9975 0.7671
nova 0 0 0.0451 0.0467 0.0015 0.046 1 1 0.9916
sylva 0.0031 0.0033 0.0065 0.0062 0.0003 0.0066 0.9998 0.9997 0.999
Overall 0.0344 0.0357 0.1117 0.1086 0.0041 0.115 0.9848 0.9846 0.9288

This entry is a complete valid challenge entry.