Can the hyper-parameter change for each new training model with each new dataset

Nguyễn Ngọc · April 16

Dear dataikuler thanks for reading my question. Hi, so my problem is when i re-train my model with different dataset (like my first dataset is from 12/10 /2024 to 12/10/2025 and my second dataset is 30/11/2024 to 30/11/2025) and then i deploy the second model i check the hyperparameter of each version and i see all of them is the same. My idea is that it should be somehow different but not in this so any advice or any tip ? Very thanks if you guys help me out :3

Alexandru · October 17

Hi,
You cannot chang this for an existing saved mode, but should be able to do this with code :
Create a new visual analysis
https://developer.dataiku.com/latest/concepts-and-examples/ml.html#the-whole-cycle
Tune the hyperparameter, compare to previous model and deploy the best model
https://developer.dataiku.com/latest/concepts-and-examples/ml.html#global-parameters-for-hyperparameter-search
Hope that helps.

Thanks

Can the hyper-parameter change for each new training model with each new dataset

Answers

Categories

Setup Info

Tags