Re:Machine learning for the impatient: algorithms tuning alg...
by gersh on Monday July 23, 2012 @02:15PM
(#4282275)
You can split it into training, validation, and test sets if you have enough data. Then, the validation set would be used for the training. Presumably by using a genetic algorithm, he isn't going for a perfect fit, which may help avoid overfitting.
[ Parent
| on HN ]
There are no comments on this comment yet.