Tuning Model Hyper-Parameters for XGBoost and Kaggle

[@lqtube]

Jeff your first 3 mins make me subcribe to your channel, feature first, esemble model second and followed by hyper parameter is such a great advice!

[@WhySoSkyHigh]

you're a legend! Well spoken, concise and informative. keep it up!

[@liquidinnovation]

Anther great video! If it's not too "easy", I'd love to see you work through the House Prices competition. The data set is small enough for us to train along with you on our laptops to merge multiple models, there are both numerical and categorical columns (so opportunities to for different feature sets into different models), and it would be a good tour of your data treatment utilities on train and test data when creating the sets (log / standardization / normalization, etc.).

[@HeatonResearch]

Thanks for the idea, I will take a look!

[@PhenomenalInitiations]

how to pass parameters to xgboost fitter with 4 dependent variables? I want to use softmax or softprob as the objective funvtion. Pls reply

[@txx568]

Excuse me，Can you share the GPU teaching of lightGbm and XGBoost package ? Thank you

[@alvaro9169]

Thanks for your time Jeff!! Very useful :). Alvaruco

[@calvinbutler5517]

Hi Jeff, thank you for your wonderful channel. Have you tried this fine tuning method in comparison to AWS Sagemaker HPO tuning?

[@ezrahouttuijnbloemendaal108]

Thanks jeff! But i don't seem to find the script you are working with. Where exactly can i find the script?

[@FelipeOliveira-pl9nu]

how to download Machine learning frameworks in Visual studio?

[@ashishvishwakarma5362]

Thanks for the video . Can you please make video on how to take advantage of Google colab GPU or TPUs .

[@HeatonResearch]

Yes that is high on my list.