WCCI Performance Prediction Challenge

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

SVM-Evidence

Submitted by Olivier Chapelle

RBF Kernel, model selection by maximizing the evidence
Performances estimated with Leave-One-Out.

Dataset Balanced Error Test guess Guess error Test score Area Under Curve
Train Valid Test Train Valid Test
ada 0.1488 0.1784 0.1869 0.1679 0.019 0.2059 0.9262 0.9119 0.8982
gina 0 0.0444 0.0696 0.0731 0.0034 0.0728 1 0.9788 0.9787
hiva 0.0399 0.3533 0.2914 0.2364 0.055 0.3465 0.9938 0.6494 0.7547
nova 0.008 0.104 0.1038 0.1137 0.0098 0.1134 0.999 0.9531 0.9536
sylva 0.0058 0.012 0.0133 0.0129 0.0004 0.0134 0.9997 0.9989 0.9984
Overall 0.0405 0.1384 0.133 0.1208 0.0175 0.1504 0.9837 0.8984 0.9167

This entry is a complete valid challenge entry.