Hey Aman. Thanks for another informative session with you...I wanted to inform you that i recently cracked Ernst & Young as a Data Analyst... but yes i still have alot to learn from brilliant minds like you and many others. Yes you have some contribution to my success so thanks alot Aman and please continue posting such beautiful videos and enlightening the masses.
Hi Aman! I love your educational videos! I was wondering if you could do a video about getting a gpu up and running in aws, or azure or google cloud? The first time that I tried it was really confusing. I think your video would be super helpful! Thanks for all the great content!
I am working on a regression problem but when i use catboostregressor and lgbmregressor for cross validation i cant get the R2 score, can only get the RMSE and MAE, can someone please tell me how to solve that?
Thank you for your video! Could you please help with 2 questions: 1) For CatBoost why do we predict y_pred based on X_test, not on pool_test? 2) Why do we use np.mean(np.sqrt(r2_score)) but not the r2_score itself?
Can a person having experience of 5 + years, in Electrical industry and telecom industry apply for a internship (if wanted to change a field ) ??? Can the internship from a small scale industries that truly provides a good expertise and working skill with knowledge of Data science be helpful in getting good job in a MNC firms ??
For Catboost decision tree grows symmetrical with >12 for both sides you mentioned. Why the same split >12 for 2 different splits and continue? Can you clear me on this line
It is to maximize the processing speed, if you break all data points on same conditions, your algorithm complexity is reduced. Catboost use this approach
Hey.... so what does it mean if I get this error R Squared for Catboost: nan Execution time: 19.085498571395874 /usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:22: RuntimeWarning: invalid value encountered in sqrt Also this is a really great tutorial. learned a lot!!!!
Hey Aman. Thanks for another informative session with you...I wanted to inform you that i recently cracked Ernst & Young as a Data Analyst... but yes i still have alot to learn from brilliant minds like you and many others. Yes you have some contribution to my success so thanks alot Aman and please continue posting such beautiful videos and enlightening the masses.
Congratulations Ishti. Good luck.
That was a fantastic video, thank you so much!
Very good Aman
Thank you.
This is an extremely important topic, very well explained..
Great video as expected
Thanks for watching Sambit. See if get catboost perform better on some other data, let me know, I m curious to see😊
@@UnfoldDataScience Sure Aman, waiting for next amazing video from you.
Excellent resource, thank you!
Hi Aman! I love your educational videos! I was wondering if you could do a video about getting a gpu up and running in aws, or azure or google cloud? The first time that I tried it was really confusing. I think your video would be super helpful! Thanks for all the great content!
CatBoost fits faster on GPU because of symmetric trees (they can be fitted in parallel). If you have terabytes of data, it's better to use CatBoost.
I am working on a regression problem but when i use catboostregressor and lgbmregressor for cross validation i cant get the R2 score, can only get the RMSE and MAE, can someone please tell me how to solve that?
Merci!
Thank you for your video! Could you please help with 2 questions: 1) For CatBoost why do we predict y_pred based on X_test, not on pool_test? 2) Why do we use np.mean(np.sqrt(r2_score)) but not the r2_score itself?
Hi, have you got the answers to the 2nd question? I am wondering the same thing. Thanks!
Hi Aman, kudos for your teaching efforts. It is easy to understand. It would if you share the this code in git or drive. thanks.
In the description
You can use a green screen to crop out the background!
Sometimes the picture becomes weird if I do that, when I move hands and all.
I m trying to find out optimum solution.
Thank you
You're welcome.
why did you apply Label Encoder to categorical (input variables )? Sklearn says , it should be used to encode target values!
683 Alberto Parkway
can you plz video on time series with the same boosting algos
Sure
@@UnfoldDataScience when will you please do a video on time series forecasting with XGboosting
Sir can you explain about quatization aware training
Ok sure sharan
Schuster Mission
can we have this code link?
like for russian google )
Thank you 😄
Maureen Manors
Can a person having experience of 5 + years, in Electrical industry and telecom industry apply for a internship (if wanted to change a field ) ???
Can the internship from a small scale industries that truly provides a good expertise and working skill with knowledge of Data science be helpful in getting good job in a MNC firms ??
First learn then apply for internship, yes it is possible to do internship and hunt for job outside.
@@UnfoldDataScience
From Where should I and can learn it from basic ?
thanks for a reply btw
Lane Parks
For Catboost decision tree grows symmetrical with >12 for both sides you mentioned. Why the same split >12 for 2 different splits and continue? Can you clear me on this line
It is to maximize the processing speed, if you break all data points on same conditions, your algorithm complexity is reduced. Catboost use this approach
@@UnfoldDataScience Thanks
7423 Davion Prairie
Hey.... so what does it mean if I get this error
R Squared for Catboost: nan
Execution time: 19.085498571395874
/usr/local/lib/python3.7/dist-packages/ipykernel_launcher.py:22: RuntimeWarning: invalid value encountered in sqrt
Also this is a really great tutorial. learned a lot!!!!
Very good Aman
Thank you