Andrea Madotto

I contacted the author of the paper [Few-shot Natural Language Generation for Task-Oriented Dialog](https://www.microsoft.com/en-us/research/uploads/prod/2020/03/2002.12328.pdf) and they provide me with the code. You can find those evaluators, for some domains at...

Hi, yes, that's a bug. This weekend I will fix it up. Thanks for letting me know Best Andrea

Hi @spolu, thanks for spotting this problem. I will check it asap.

Hi, you are right, this should be fixed. Thanks

Hi, yah, I tried again, and I cannot reproduce the same results. I think I used different hyper-parameter. I did not sync what I had in my server, so I...

Hi, if you want an interactive script, please refer to the [AdapterBot Demo](http://adapterbot.emos.ai/) and [AdapterBot Repo](https://github.com/HLTCHKUST/adapterbot) where we release a better interactive script. Check also this issue for more information:...

+1

Hi, in this repo, we didn't push the training script. However, for training the adapter you can refer to two of our previous repositories: https://github.com/andreamad8/PPCM and https://github.com/zlinao/VGLM. Hopefully, we are...

Any plan to add this? one easy solution that seems not to be working could be: ``` os.environ["WORLD_SIZE"] = str(os.environ["WORLD_SIZE"] // model_parallel_size) os.environ["RANK"] = str(os.environ["RANK"] // model_parallel_size) ``` I tried,...