What about a demo where one can do classification with different methods (That are all pre-configured somehow) But then have a chain of different things.
Preprocessor choices (PCA, remove mean, whitening, etc)
Classifier (GP, SVM, Tree) all with fixed or automagic parameter selection
visualisation of result
evaluation of performance measure of choice (accuracy, AUC, etc)
@Saurabh7
What about a demo where one can do classification with different methods (That are all pre-configured somehow) But then have a chain of different things.
Preprocessor choices (PCA, remove mean, whitening, etc) Classifier (GP, SVM, Tree) all with fixed or automagic parameter selection visualisation of result evaluation of performance measure of choice (accuracy, AUC, etc)