How to install xgboost package on x32 bit windows for python?

xgboost
python

#1

Hi,

I am using anaconda and need to install xgboost package, as it is not preinstalled.I tried method described on github: https://github.com/dmlc/xgboost/tree/master/windows, with no success. I need to install for x32 bit windows 7. Any help would be appreciated.


#2

@munitech4u Unfortunately an xgboost package for Windows does not appear to be available in the Anaconda repo.

>>anaconda search -t conda xgboost
Using anaconda-server api site https://api.anaconda.org
Run 'anaconda show <USER/PACKAGE>' to get more details:
Packages:
     Name                      |  Version | Package Types   | Platforms
     ------------------------- |   ------ | --------------- | ---------------
     akode/xgboost             |      0.3 | conda           | osx-64
     aterrel/xgboost           |    0.4.0 | conda           | linux-64, osx-64
                                          : An optimized general purpose gradient boosting library.
Found 2 packages

#3

Yes, it is not available with anaconda. Infact the direct pip install is disabled for this package, as it has some problem with windows in that kind of installation mode. So i referred to the the github link, where the method to compile and install is given on windows platform. I only had partial success with that and had posted the error log in github. I was wondering if somebody has installed it on windows or might have pre-compiled libraries.


#4

You can clone the XGBoost directly from github repo.

Below is the link to the the steps needed to make it working on windows:

Hope this helps.


#5

Hi…

Have you installed xgboost in python.
Please guide how to install. I am stuck on it for quite few days.
Thanks…


#6

Hi,

Any one installed xgboost in anaconda ?
could you please suggest the detail step for new one…

Thanks,
venky


#7

so far no success, there is still some problem in compilation, I think they are working on it. For now, rely on R for running xgboost


#8

Guys,
try this method. this is the best method i have seen so far (Python3)(dont have to compile it yourself).
will work on anaconda, and has scikit-learn wrapper.

1.download xgboost whl file from http://www.lfd.uci.edu/~gohlke/pythonlibs/ (make sure to match your python version and system architecture, e.g. “xgboost-0.6-cp35-cp35m-win_amd64.whl” for python 3.5 on 64-bit machine)
2. open command prompt
3. cd to your Downloads folder (or wherever you saved the whl file)
4. pip install xgboost-0.6-cp35-cp35m-win_amd64.whl (or whatever your whl file is named)

If you find it won’t install because of a missing dependency, download and install the dependency first and retry.

If it complains about access permissions, try opening your command prompt as Administrator and retry.

I found this on stackoverflow and works well in my system


#9

About 18 months after my previous reply, I have some good news (for those who may not be aware of this): xgboost is available from Anaconda. All major OSs and architectures are covered. For both Python and R. This is not a third-party package on Anaconda Cloud but one that is maintained by Continuum Analytics, the people behind Anaconda.

https://anaconda.org/anaconda/py-xgboost
https://anaconda.org/r/r-xgboost