Question about supporting multi variable anomaly detection

I’m also new to HTM and I also intend to do the same.
I think we should implement a custom encoder that takes the two varyables as input in order to do that. Take a look at the CoordinateEncoder and try to figure it out. In order to implement a custom Encoder you should inherit from Encoder class.

Found this topic on writing your own custom Encoder: [Writing a custom encoder](http://Writing a custom encoder)

Bellow is a scheleton of a custom Encoder automatically generated by Netbeans:

/*
 * To change this license header, choose License Headers in Project Properties.
 * To change this template file, choose Tools | Templates
 * and open the template in the editor.
 */
package org.numenta.nupic.examples.napi.hotgym;

import java.util.List;
import org.numenta.nupic.encoders.Encoder;

/**
 *
 * @author My Self
 */
public class MyCustomEncoder extends Encoder{

    @Override
    public int getWidth() {
        throw new UnsupportedOperationException("Not supported yet."); //To change body of generated methods, choose Tools | Templates.
    }

    @Override
    public boolean isDelta() {
        throw new UnsupportedOperationException("Not supported yet."); //To change body of generated methods, choose Tools | Templates.
    }

    @Override
    public void encodeIntoArray(Object inputData, int[] output) {
        throw new UnsupportedOperationException("Not supported yet."); //To change body of generated methods, choose Tools | Templates.
    }

    @Override
    public List getBucketValues(Class returnType) {
        throw new UnsupportedOperationException("Not supported yet."); //To change body of generated methods, choose Tools | Templates.
    }

    
    
}
1 Like