[XGBoost] Is Anyone Using XGBoost?


79
views
0
11 weeks ago by

Hi All,

I anyone using XGBoost as a machine learning algorithm?  I am looking for ideas for tuning that may be more efficient than a brute force search through the many parameters.

Thanks,

Kevin

add commentfollow this post modified 10 weeks ago by Kevin Maher   • written 11 weeks ago by Kevin Maher  

2 Answers


0
11 weeks ago by

Hi Kevin,

I'm planning to use xgboost too, but I'm not quite to that step yet in my project. I'm  still working on feature extraction stuff with my data set.

I have never used it before, but am looking forward to trying it out :)

-Aubrey

 

add comment written 11 weeks ago by Aubrey Sharwarko  

I should have said, but I'm using the Python version. Hoping to get it going by Monday.

written 11 weeks ago by Aubrey Sharwarko  
0
11 weeks ago by

Hi Kevin.

I haven't settled on a final algorithm choice yet, but I am having a lot of success with GBM's at the moment. However, I did find this resource last semester while reading about XGBoost in R:

https://www.hackerearth.com/practice/machine-learning/machine-learning-algorithms/beginners-tutorial-on-xgboost-parameter-tuning-r/tutorial/

Most of the tuning parameters are listed and helpful tips are provided for starting values, things to look out for, etc...

I found it very useful while tuning in R, but the same tuning concepts could be applied to Python as well. 

add comment written 11 weeks ago by Justin Moore  

Hi Justin,

Thanks for the link.  As far as I can tell though the only way to tune it is an exhaustive grid search.

Regards,

Kevin

written 10 weeks ago by Kevin Maher  
Please log in to add an answer/comment or follow this question.

Similar posts:
Search »