Forecast Metrics For Foreach Processor

Modified on Tue, 30 Nov 2021 at 11:44 AM


This processor is used to calculate different error measures from forecasts generated in a Foreach branch (created by the Foreach Processor). It outputs a single-row dataset containing the forecast metrics foreach distinct value in the dataset.

NOTE THAT: It is very recommended to have a basic understanding of the functionality of "For Each" as well as the "Forecast Metrics" processors.


This processor requires a Dataset that has been processed by a "For Each" processor. Additionally this Dataset should contain a forecast column and an original value column.


This processor should be linked to the output node of a For Each Processor. The latter should be linked to the output node of a Forecasting processor (such as Decision Tree Regression Forecast).

An example of a Workflow pipeline looks as follows:

When opening the configuration menu of this processor, the following interface shows up:

NOTE THAT: The same column in the For Each and in the Forecast Metrics For Foreach (third configuration field) processors should be selected.


This processor provides two output nodes:

  • Left Node: returns the input Dataset grouped with respect to the values of the selected column in the Foreach processor (a drop-down menu is shown in the result table)
  • Right Node: return the calculated error metrics Mean Squared Error (MSE), Root Mean Squared Error (RMSE) and Mean Average Error (MAE) grouped by the values of the same column selected in the Foreach processor (there is also a drop-down menu in the result table)
The output of the right node can help track the forecasting errors with respect to specific values of the selected column which can help understand correlation between this column and the column on which forecasting is performed on.Furthermore this can help detect Outliers.


In the following Workflow we used Custom Input Table to create a toy Dataset, Horizontal Split Processor to divide it into training and test sets which both will be fed to a Decision Tree Regression Forecast which output will be linked to a Result Table (direct visualization of the forecasting processor output). The output is linked to a Forecast Metrics Processor and a Foreach Processor. We then insert the Forecast Metrics For Foreach Processor.

The workflow looks as follows:

Please refer to the attachment to follow the Workflow execution.

Related Articles

Decision Tree Regression Forecast

Decision Tree Classification Forecast

Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select atleast one of the reasons

Feedback sent

We appreciate your effort and will try to fix the article