Topic 10. Part 2. Key ideas behind Xgboost, LightGBM, and CatBoost. Practice with LightGBM

In this part, we discuss key difference between Xgboost, LightGBM, and CatBoost.
Practice with logit, RF, and LightGBM - www.kaggle.com...
Main site - mlcourse.ai
Kaggle Dataset - www.kaggle.com...
GitHub repo - github.com/Yor...

[@abhishek-shrm]

Great video. I learned a lot from it.

[@rrrprogram8667]

Nice video.. Thanks for making

[@zes3813]

no such hting as tpix or not

[@historia_tego_swetra]

hm, scratching my head for weeks now. like howard said, what's the point of predicting churn?

[@festline]

Do you want to argue that predicting churn is useless? And all these analysts are wasting their company's money?

[@historia_tego_swetra]

@@festline okay, I predicted a churned user X, how is it going to help me? Again, as the majority of analysts out there is going to tell, such a user will cost x+times more to keep hence what's the point in 'predicting'it? the conventional wisdom says it makes sense to harden things that keep people from leaving... and that's not even my opinion, it is out there :D

[@festline]

I didn't work directly with churn, so I don't have arguments better than predicting those who'd churn and making a nice proposal to them. All telecom operators do exactly this. At the same time, I wonder how Howard motivates his position. Did he work with churn? I guess no.

[@historia_tego_swetra]

@@festline Jeremy Howard? :D I bet he did