microsoft
is the final model fitted to the whole train set?
Dear, first of all thanks a lot and compliment! Then, in documentation is not clear one thing: after a fit the final model is fitted to the whole train set? I need to do something else in order to fit the model to all data and go in production?
thanks in advance and waiting your kindly reply...
luigi
for example, verstack has fit(X,y) to "Execute LGBM model hyperparameters tuning " and has fti_optimized(X,y) to "Train model with tuned params on whole train data"..............
waiting Your help....
Hi @luigif2000,
Thank you for the question and sorry for the late response. Yes, the model is fitted on the full train set. The relevant keyword argument in the fit function is retrain_full. You can find more details in the docstr for retrain_full.
Please let me know if your question is not resolved. Thank you!
ok thank's so much....
you are very welcome
best regards
luigi
Il giorno mar 2 ago 2022 alle ore 23:14 Qingyun Wu @.***> ha scritto:
Hi @luigif2000 https://github.com/luigif2000,
Thank you for the question and sorry for the late response. Yes, the model is fitted on the full train set. The relevant keyword argument in the fit function is retrain_full https://github.com/microsoft/FLAML/blob/main/flaml/automl.py#L570. You can find more details in the docstr for retrain_full.
Please let me know if your question is not resolved. Thank you!
— Reply to this email directly, view it on GitHub https://github.com/microsoft/FLAML/issues/650#issuecomment-1203220514, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABNWP4MGGORXWZLGPC34A6TVXGFTBANCNFSM54ZZDG4Q . You are receiving this because you were mentioned.Message ID: @.***>
You are welcome. Let me know if you need more help! Do you mind sharing what kind of production you are working on and using FLAML? And can you please fill out this survey to share your feedback about FLAML: https://docs.google.com/forms/d/e/1FAIpQLSd0u4BbSzr-m9hgM8rmPlYB_JJoFXeHhuoXI9YFbHDRWrDlOQ/viewform?usp=sf_link
Your feedback is appreciated. Thank you!
sure, i did...no more help for now...have a nice day thanks again
luigi
Il giorno ven 5 ago 2022 alle ore 13:58 Qingyun Wu @.***> ha scritto:
You are welcome. Let me know if you need more help! Do you mind sharing what kind of production you are working on and using FLAML? And can you please fill out this survey to share your feedback about FLAML:
https://docs.google.com/forms/d/e/1FAIpQLSd0u4BbSzr-m9hgM8rmPlYB_JJoFXeHhuoXI9YFbHDRWrDlOQ/viewform?usp=sf_link
Your feedback is appreciated. Thank you!
— Reply to this email directly, view it on GitHub https://github.com/microsoft/FLAML/issues/650#issuecomment-1206365374, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABNWP4KNVWCQL4O67T4V5S3VXT6VHANCNFSM54ZZDG4Q . You are receiving this because you were mentioned.Message ID: @.***>