[PyMVA] Refactoring of sklearn classifier with multi-class capability, unit-testing and ranking #343
Conversation
|
Can one of the admins verify this patch? |
| ////////////////////////////////////////////////////////////////////////// | ||
|
|
||
| #ifndef ROOT_TMVA_PyMethodBase |
There was a problem hiding this comment.
Could you remove these redundant header guards? We just removed all of them from the current repo in #341
There was a problem hiding this comment.
What is the current coding style? Moving all the guards into the headers? What about using #pragma once? That would be nice ;)
There was a problem hiding this comment.
I'll fix it later!
There was a problem hiding this comment.
Not sure about all the details of the coding style, but at least here I think it should be normal header guards in the respective header and no "double guards"around the include :)
You are correct. We just changed the coding convention to not longer require the 'double guard' (which is nowadays redundant, in the past it saved in disk access). Cheers, |
|
@phsft-bot build! |
|
Can one of the admins verify this patch? |
|
Hi Stefan, I would like to merge this PR. Can you please verify the shown conflict with the master branch ? |
…dependent from sklearn
|
I've rebased and solved the conflict! |
|
Thank you. It is now merged |
This is a rebase and combination of the PRs #303 #314 #315.
GetMvaValuesimplementation from base class to method to stay independent from sklearnYou can test the changes with
ctest -V -R PyMVA.