Can you provide the data with unprocessed code for the 28563 examples?

[Jun-jie-Huang]

Hi, thanks for your wonderful dataset and repo.

I'm developing a model for code summarization and want to take notebookCDG as one of my tasks. But I couldn't find the original dataset.

I follow the instructions in README. I download the data from your link the fully processed data (as a pkl file) can be downloaded [HERE](https://ibm.biz/Bdfpk6) and also checked the data in huggingface dataset. I only find the data with processed code string in code.seq file. But the punctuation marks are removed.

So could you provide the data with original code without any tokenization or removing punctuations for search? I'd appreciate it if you could release the dataset!

Thanks,
Junjie

[xuyeliu]

Readme already update the raw notebook and raw pair