The final training is performed on the combined training and validation dataset. There are three different training functions, one for each type of classifier:
- feature engineering methods:
trainRowFeatureMethod - 1D CNNs:
trainRowDataMethod - 2D CNNs:
trainImageMethod
The training functions take the classifier name as a string:
trainRowFeatureMethod("AdaBoost");
trainImageMethod("CNN2d3Layer");