Classification ID instead of regression value in MLP

Hi all,

I want to do a classification with a RProp MLP Learner and Predictor. As Input i use different RSS Signals from wlan access points and as target i use a id (represents an x and y position).

After training und predicition i get no classes (represented as id) as result instead i get a regression value. How can i handle this issue that i get predicted id for several rss inputs, instead of the predicited regression values. What is my fault?

Thanks for your help

Marcus

Hi Marcus,

The MLP Predictor node contains the final prediction in a column called PredClass. The different numeric values are representing the probability for each class, the names start with Cluster_1 (Neuron_1) aso.

Regards, Thomas

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.