some whacked out stuff is going on here

[samkeene]

clear all

ds = prtDataGenNoisySinc;

pca = prtPreProcPca('nComponents', 25);

pca = pca.train(ds);

pca = pca.kfolds(ds,2);

pca = pca.train(ds);

[newfolder]

Nothing wacked out that I can see. The line

pca = pca.kfolds(ds,2);

assigns the output dataset of the kfolds method to the variable pca.

The next line assumes pca should be a prtPreProcPca object, which it no longer is, and then it errors.

Am I missing something?

[samkeene]

oh yea, that last line is retarded, forget that.

i was trying to come up with a simple example of something else. i failed. i think my real issue is PCA and regression objects. i'll ask someone tomorrow