Differences between BugFixTokenPairs files

[dangnguyenngochai]

In the data folder, there are two types of datasets, one is named BugFixNoDup_ and BugFixTokenPairs_commits. Can you explain the difference between these two types of datasets? If I want to re-run the experiment for pre-training on BugFix data then which dataset should I be using.

[chenzimin]

I believe that we used BugFixNoDup_ to generate the dataset. But if you are interested in using the dataset with context size 3, they are already in the data folder of each trained model.

[dangnguyenngochai]

Do they have any different with the *_commit dataset ?

…

On Wed, May 4, 2022 at 8:03 PM chenzimin ***@***.***> wrote: I believe that we used BugFixNoDup_ to generate the dataset. But if you are interested in using the dataset with context size 3, they are already in the data folder of each trained model. — Reply to this email directly, view it on GitHub <#9 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQJ6JV73CAJJODDX6SMDR2LVIJYTJANCNFSM5U7PVMGA> . You are receiving this because you authored the thread.Message ID: ***@***.***>