search

kaz-Anova / StackNet

StackNet is a computational, scalable and analytical Meta modelling framework
MIT License
1.32k stars 344 forks source link

What if I want to use xgboost as first layer model? #9

Closed tcsere07 closed 6 years ago

tcsere07 commented 7 years ago

If I want to use xgboost as first layer model, what should I do?

kaz-Anova commented 7 years ago

Me too :)

I am still trying to integrate that . You have to w8 for it or try to add meta predictions with xgboost to your training data as in here in the script create_files_v1.py . On a positive note I am expecting to add this this week - if everything goes as planned. But it will be fairly untested :(

tcsere07 commented 7 years ago

Thx! I'm looking forward to seeing this new feature. Another question:How could I extract the out-of-fold predictions and the test prediction of first layers?Because I want to use them to train a second layer model, like xgb, lgb, keras etc...

kaz-Anova commented 6 years ago

Xgboost has been added. I am closing this.