Closed jlmeunier closed 6 years ago
Hi @jlmeunier, thanks a lot for this! Sorry for not merging your older PR when I made the recent release, I wanted to have it out quickly before my paper presentation. I will get on reviewing this soon, I am travelling over the next couple of days but I should be able to look next week.
For reference, I will probably be taking over the "release engineering" part of AD3 since I will be working very closely with it in the future, but of course @andre-martins remains the boss.
Just checking: this supersedes #16 which we can close now, right?
Hi Vlad, Thanks for your answer! Yes, the latest PR supersedes #16, which I closed. (Didn't know about closing PR before your post...) JL
Thanks a lot for contributing @jlmeunier !
@vene re the Xerox copyright: yes, it doesn't seem to make much sense to have it here. I guess this will disappear when the README.md file gets merged with the official README?
Hi Vene and André, Thanks for having looked at this PR! I will do my duty ASAP, but I'm too busy those days. So, expect some action from me in the coming weeks, or by end of year at the latest.
Hi Vene, I did what was needed (PEP8 compliance, example of use of the Python wrapper, explanation of its API) Have a look and let me know. Thanks (and wishing you an happy Xmas!) JL
(And yes, no need to propagate the Xerox copyright.)
Hi, thanks a lot for the update! I promise to take a look soon.
Hi Vene sorry, while testing with Python3, a bug that I introduced when formatting the code for PEP8 showed up... So now, things are fine.
Merged via #20, thanks @jlmeunier!
Hi, I'm proposing an extension to the AD3 Python wrapper, to exploit the AD3 logic constraint functions. Also, AD3 supports nodes having a variable number of states. This is another extension to the wrapper. Those 2 features have been extensively tested and used in an extension to pystruct, which I proposed to Andreas. All this work is reported in a CAp 2017 paper ( arXiv:1708.07644) and ICDAR OST paper to come. Thanks JL
Extensions: