The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here
Bagging - how to see averaged predictions?
Learner III
Dear Forum,
I apologise in advance for the newbie question, but I cannot seem to figure this one out.
I have a simple regression problem and which i would like to solve with a neural network ensemble generated by bagging. So I took the bagging operator, stuck the neural network in, adjusted my parameters and then chose the "Apply Model" operator. I could now use the Performance operator and average the errors, however, I would like to see the averaged predictions. At the output port of "Apply Model", I can only see the meta-data of the predictions, but I cannot seem to find a way to average them and display.
Any ideas and comments are very appreciated.
Thank you very much,
Jannnikks
I apologise in advance for the newbie question, but I cannot seem to figure this one out.
I have a simple regression problem and which i would like to solve with a neural network ensemble generated by bagging. So I took the bagging operator, stuck the neural network in, adjusted my parameters and then chose the "Apply Model" operator. I could now use the Performance operator and average the errors, however, I would like to see the averaged predictions. At the output port of "Apply Model", I can only see the meta-data of the predictions, but I cannot seem to find a way to average them and display.
Any ideas and comments are very appreciated.
Thank you very much,
Jannnikks
0
Answers
the predictions of the incorporated neural nets are averaged by the Bagging Model itself. You will only see the average in the prediction column of your apply data set. To see this predictions before they are accumulated in the performance vector right click on the apply model operator and select "breakpoint after". Then the process stops after executing this operator and the results on the connected outputs are shown.
For more details about the program and it's gui please refer to the manual.
Greetings,
Sebastian