Error in x validation loop end?

KNIME Analytics Platform
1

Hi. I want to use the x-validation loop with a neronal network.

I figured out that the mean squared errors are calculated in the X-Aggregator node as followed (directly copied from the code):

for (DataRow row : in) {
                RowKey key = row.getKey();
                DoubleValue target = (DoubleValue)row.getCell(targetColIndex);
                DoubleValue predict = (DoubleValue)row.getCell(predictColIndex);
                double d = (target.getDoubleValue() - predict.getDoubleValue());
                errorSum += d * d;
                r++;

                m_predictionTable.addRowToTable(row);
                subExec.setProgress(r / (double)rowCount, "Calculating output "
                        + r + "/" + rowCount + " (\"" + key + "\")");
                subExec.checkCanceled();
            }

            errorSum = Math.sqrt(errorSum);

 

As I see it it is wrong. Isn't the mean sqaure error?:

mean(y(expected)-y(predicted)) * (y(expected)-y(predicted)) )

Here they calculate:

(sqrt(sum of all(y(expected)-y(predicted)) * (y(expected)-y(predicted)) )) ) /number of datasets

or is there a sense? First I thought it is the root mean square error.

But the in the root mean square error the root is calculated after the mean.

2

The code seems wrong to me, too.

 

The last line should be:

errorSum = Math.sqrt(errorSum/r); // note check for division by zero!

 

I'm new to KNIME and therefore don't know how to report a bug or just fix the code myself and check it in...

3

This has already been fixed some months ago.

4