CV impementation about xgboost HOT 6 CLOSED

Thank you for choosing XGBoost frist. You can easily write down CV code using Python interface.

from xgboost.

May I know your comments about CV estimation as given in
https://www.kaggle.com/blobs/download/forum-message-attachment-files/1290/test_xgboost.py
?

On Sat, Aug 23, 2014 at 5:34 PM, Bing Xu [email protected] wrote:

Thank you for choosing XGBoost frist. You can easily write down CV code
using Python interface.

—
Reply to this email directly or view it on GitHub
https://github.com/tqchen/xgboost/issues/33#issuecomment-53151010.

from xgboost.

Cosmos,
There are many ways of CV - it is not always n-fold Cross-validation. Imagine time series data - you will need time based cross validation. you have to use wrappers around xgboost for that.....Let xgboost be what it is - a state of the art ML algorithm that is going to be the best Gradient boosting tree based qualifier out there

from xgboost.

see an experimental version of CV and examples here: https://github.com/tqchen/xgboost/blob/master/demo/guide-python/cross_validation.py

from xgboost.

and example of higgs cv, provided by Bing Xu, is in https://github.com/tqchen/xgboost/blob/master/demo/kaggle-higgs/higgs-cv.py

from xgboost.

Thanks a lot for the links.

On Thu, Sep 4, 2014 at 11:53 PM, Tianqi Chen [email protected]
wrote:

and example of higgs cv, provided by Bing Xu, is in
https://github.com/tqchen/xgboost/blob/master/demo/kaggle-higgs/higgs-cv.py

—
Reply to this email directly or view it on GitHub
https://github.com/tqchen/xgboost/issues/33#issuecomment-54522141.

from xgboost.