I am at the moment using prediction.strength{fpc} to test how well my kmeans
clustered data classifies using knn.
I do that by calling this function.
prediction.strength(training, Gmax = 10, M = 5, classification = "knn",
count = TRUE,nnk = 20)
Is it possible to provide a separate test set to compute the mean correct
prediction = mean.pred, instead of using the training set itself... Or to
compute it from some form of cross validation.
--
View this message in context:
http://r.789695.n4.nabble.com/prediction-strength-test-prediction-strength-with-an-separate-test-set-tp4719188.html
Sent from the datatable-help mailing list archive at Nabble.com.
_______________________________________________
datatable-help mailing list
[email protected]
https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/datatable-help
