Jason Wu

@CaoHoangTung, The dialogue_acts.json has the system act annotations from the original dataset. "1" means the annotation of the first system utterance.

Hi, what was the command you ran? Please check these [few lines](https://github.com/jasonwu0731/ToD-BERT/blob/master/main.py#L46) to make sure and print what you have for your unified_meta.

The problem here is that some datasets you used do not have the DST labels to be trained as defined similarly to MultiWOZ. You can first run only using Multiwoz...

@Bosheng2020 Can you provide the full command you were running?

@Bosheng2020 Are you trying to run a pretraining task or DST task? If the pretraining task, you need to use task=usdl with [run_tod_lm_pretraining.sh](https://github.com/jasonwu0731/ToD-BERT/blob/master/run_tod_lm_pretraining.sh#L15) If DST task, you need to run...

@aliyabannaeva Can you check if you can run [this command](https://github.com/jasonwu0731/ToD-BERT/blob/master/evaluation_pipeline.sh#L31)? If not, please copy and paste the error message here, thanks.

Hi, Can you provide what is the error when you run the RCL training? We did not focus too much on parallel training at that time and used the huggingface...

Perplexity of a dataset could be helpful. Thank you!

@shmsw25 Thank you for your response! After I change that line, I got the following error. ``` Caused by op u'towers/gpu_0/loss/ans_loss/SparseSoftmaxCrossEntropyWithLogits_4/SparseSoftmaxCrossEntropyWithLogits', defined at: File "/home/jason/anaconda2/lib/python2.7/runpy.py", line 174, in _run_module_as_main "__main__",...

@mhmda yes, you are right. Should change to 4