search

microsoft / FLAML

A fast library for AutoML and tuning. Join our Discord: https://discord.gg/Cppx2vSPVP.
https://microsoft.github.io/FLAML/
MIT License
3.91k stars 508 forks source link

Question: Why not retrain on full dataset if eval_method = cv? #131

Closed stepthom closed 3 years ago

stepthom commented 3 years ago

I noticed that FLAML will only retrain on the full dataset if the eval_method parameter is set to 'holdout':

https://github.com/microsoft/FLAML/blob/b04b00dc9d4bc47a4135f777ca7e419854ceb140/flaml/automl.py#L910-L911

Why not retrain on full dataset for other eval_methods, such as 'cv'?

sonichi commented 3 years ago

The retraining happens right after cross-validation: https://github.com/microsoft/FLAML/blob/b04b00dc9d4bc47a4135f777ca7e419854ceb140/flaml/ml.py#L259

stepthom commented 3 years ago

Oh, I see - thank you very much! 👍