Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[python3] use python3 dmlc tracker in rabit tests #525

Merged
merged 1 commit into from
Apr 19, 2019
Merged

[python3] use python3 dmlc tracker in rabit tests #525

merged 1 commit into from
Apr 19, 2019

Conversation

chenqin
Copy link
Contributor

@chenqin chenqin commented Apr 17, 2019

upgrade to dmlc-submit to python3 as alternative to #524

@chenqin chenqin marked this pull request as ready for review April 17, 2019 02:21
@chenqin
Copy link
Contributor Author

chenqin commented Apr 17, 2019

dmlc/xgboost#4379

@CodingCat
Copy link
Member

how we verify python3 can make rabit test pass?

@chenqin
Copy link
Contributor Author

chenqin commented Apr 18, 2019

I tested in my local OS X and Linux, to have Travis chi working we need to get either this or deadlock or in and git update sub module

@CodingCat
Copy link
Member

as the common practice like before, we should point rabit to the private branch having python 3 to verify

and

there are still bunch of places in the code base using python instead of python3, we'd fix in one shot instead of leaving code base in an inconsistent status using py2 and py3 simultaneously

@chenqin
Copy link
Contributor Author

chenqin commented Apr 18, 2019

This is getting convoluted, if we need to fix rabit problem do we need to wait all xgboost upgrade to python3? Otherwise, we should merge #524 and keep upgrade python3 in parallel

@CodingCat
Copy link
Member

if we need to fix rabit problem do we need to wait all xgboost upgrade to python3?

make a simple search, https://github.com/dmlc/dmlc-core/search?l=Python&q=python

Otherwise

otherwise against what?

we should merge #524

alternatives as highlighted

image

@chenqin
Copy link
Contributor Author

chenqin commented Apr 18, 2019

524 assume we support python2 and python3
525 meant to drop support of python2 and assume deadlock issue go away with python2

unless we can assume https://github.com/dmlc/xgboost/tree/master/python-package/xgboost works with python3, python3 upgrade might be more than we expected. Ultimately, my impression is we should get away all python2 packages from pip install and move to pip3.

@CodingCat
Copy link
Member

unless we can assume https://github.com/dmlc/xgboost/tree/master/python-package/xgboost works with python3, python3 upgrade might be more than we expected.

it is not related to what we are talking about here, and if you have any concern or you think you find some cases XGBoost cannot move forward to py3, please stated in dmlc/xgboost#4379

Ultimately, my impression is we should get away all python2 packages from pip install and move to pip3.

yes, that's why I said, if you want to upgrade to python 3 in this code base, you'd upgrade everything appeared in https://github.com/dmlc/dmlc-core/search?l=Python&q=python

@hcho3
Copy link
Contributor

hcho3 commented Apr 18, 2019

@chenqin FYI, XGBoost has been tested with Python 3 for a good while.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants