How to update a regression model on new data without retraining it in full?

I built a medium tree model in the Regression Learner App and exported it to the Workspace.

Is it possible to update the model with new data without training the whole data again?

Answer 1

[Harsha Priya Daggubati]

Hi,

I think this thread might help you:

https://www.mathworks.com/matlabcentral/answers/481300-how-to-re-train-a-model-optimized-by-bayesian-optimization-on-new-data?s_tid=answers_rc1-2_p2_MLT

[Christoph]

The solution from the thread works for fitting new data but it overwrites the old data set. Is there any way to keep the old data set and only update the model with new data.

The problem is that i have a lot of data to train and i'm trying to do it piece by piece.

[Harsha Priya Daggubati]

I guess the thread suggests creating a template of your already fitted model and using it to new data.

Also, this approach might help you too.

https://www.mathworks.com/matlabcentral/answers/264160-how-to-save-and-reuse-a-trained-neural-network