requirements for imdb dataset about multibench

[mrbeann]

BTW, I also found that humor and sarcasm not work properly.
e.g., for sarcasm

    alldata['train'] = drop_entry(alldata['train'])
KeyError: 'train'

[Vanvan2017]

So sorry for the error in sarcasm and humor, this is may due to the dictionary of the processed data not updated to the unique data_loader for all when refactoring the code, please check this version: https://drive.google.com/drive/folders/1NTtr27Of3h7ou03Qvt51wYgf9hTMJ4YL?usp=sharing, this should work fine~

[Vanvan2017]

BTW, I also found that humor and sarcasm not work properly. e.g., for sarcasm

    alldata['train'] = drop_entry(alldata['train'])
KeyError: 'train'

We also have updated data and new features of data loaders in README~

[mrbeann]

Thank you for your reply, what about the IMDB dataset?

[neal-ztwu]

Thank you for your reply, what about the IMDB dataset?

Sorry for the late reply.
I am not sure why your situation happened but I have pushed an environment file in the repo. Could you please try again using the same environment with our work? Thanks!

[mrbeann]

Thank you for your reply, what about the IMDB dataset?

Sorry for the late reply. I am not sure why your situation happened but I have pushed an environment file in the repo. Could you please try again using the same environment with our work? Thanks!

Thank you for your reply. It still does not work. It reported the "block" is missing.

BTW, recent updates seem to make this repo unable to run. @Vanvan2017 @neal-ztwu