language_modeling_via_stochastic_processes

language_modeling_via_stochastic_processes copied to clipboard

question of a recent commit (and also reproducibility questions)

2

Hi Rose, The recent commit changed x_tp1=x_tp to x_tp=x_tp1 (https://github.com/rosewang2008/language_modeling_via_stochastic_processes/commit/cb3d3454433d821c606bc224d42ee81b7cd3754f?diff=split?), but are your results based on this new commit or the old one? Since I also got different numbers on...

da03

Issue in data loading

1

It seems to me that this line should be changed to `if 'tm' in self.name` (https://github.com/rosewang2008/language_modeling_via_stochastic_processes/blob/5cbc3eed581eba6444c471bfe716bd56db0f5253/language_modeling_via_stochastic_processes/transformers/src/transformers/data/datasets/language_modeling.py#L1201), since you were using `self.start_conversation` and `self.end_conversation` to split the training and test sets...

da03

code for roc stories?

1

It seems that ROC Stories use a different setup text infilling, but I can't find code in this repository for infilling. Am I missing anything or is the code not...

da03

https://github.com/rosewang2008/language_modeling_via_stochastic_processes/blob/5cbc3eed581eba6444c471bfe716bd56db0f5253/language_modeling_via_stochastic_processes/src/datasets/wikihow.py#L41 Seems that there's an extra space here, which would result in not splitting sentences.

da03

All .sh files in encoder/decoder sub-folders use the following command: `nlprun -n .....` Which package needs to be installed for the **nlprun** command? This is not mentioned in the Setup...

sindhu-pr