Support for Chainer 2.0

[Sree Harsha Ramesh]

There are issues running the current build, due to code-breaking changes introduced in Chainer version 2.0. We should specify the chainer version during the installtion process as 1.24.0 or make changes to the existing code to support v2.

@aonotas, Let me know if you need me to send a pull request.

[himkt]

hi @harsha-surukam and @aonotas, I have edited deep-crf for chainer v2 support last week.
so I make the pull request.
please tell me if there are something wrong... (I am newbie to chainer.

[Sree Harsha Ramesh]

@himkt I guess, you have only commented out the problematic parts. This definitely works, but the better solution imho would be to upgrade from v1 to v2 according to the guide.

[himkt]

@harsha-surukam thanks for reply 😃
I understand that it needs to follow the upgrade guide, I close the pull request.

[Motoki Sato]

@harsha-surukam Thank you for your comments!
If you have fixed deep-crf for v2 support, please send me PR.

[Motoki Sato]

I think we need fix deep-crf code following points:

• train argument in dropout
• volatile argument in Variable
• use_cudnn argument in L.NStepBiLSTM

Should we switch code according to chainer.__version__?

[Sree Harsha Ramesh]

Yes, I think that should help identify the installed version of chainer. But, I'm not sure if it's a good idea to have two branches in the code, based on the version of a library. Is it the best practice?

Or, should we just enforce people to upgrade chainer? Because, currently the default chainer version installed is 2.0.

[Motoki Sato]

I asked Chainer developer just now.
He told me that we need to branch code based on chainer.__version__.

[Sree Harsha Ramesh]

Cool. Let me send you a PR in a bit.

[Motoki Sato]

Thank you for your help!!

[Motoki Sato]

Hi @harsha-surukam

Now deep-crf can run on v2.0 and v1.24.0!