Hello everybody, I'm new to R. I'm trying to set up a Rbf network using the rbftrain function but I get an error I can't understand. this is what I type: rbftrain(input,3,output,visual=F) where input is a matrix with 198 rows and 6 columns and output is a matrix with 198 rows and one column. Both matrices are missing values free. This is what I get from R: Error in kmeans(inp[, i], neurons) : more cluster centers than distinct data points. I don't understand this. Moreover, I can actually train a rbf network with 2 neurons but not with 3 neurons. I'm sure I have far more than 3 distinct data point both in "input" and in "output" so I don't get the meaning. Can you please help me? Thank you so much everybody. Invernomuto -- View this message in context: http://www.nabble.com/help-needed-with-rbftrain-tp17161948p17161948.html Sent from the R help mailing list archive at Nabble.com.