326 Commits

Author SHA1 Message Date
8360b8bbe3 added small sh script that i forgot to add when i
created it...
2025-06-28 11:09:47 +02:00
c6650b2b87 made code more consistent 2025-06-23 11:19:36 +02:00
beaef7310f updated readme 2025-06-12 17:31:13 +02:00
9879c9e528 updated readme 2025-06-12 17:30:58 +02:00
42f02595f0 updated readme 2025-06-12 17:29:49 +02:00
c52261560f updated requirements 2025-06-12 17:23:37 +02:00
7bb9eda8a4 removed useless files 2025-06-12 17:23:11 +02:00
3e98eba9d2 updated readme 2025-06-12 17:18:45 +02:00
31402752c8 updated readme 2025-06-12 17:16:06 +02:00
f898028fcc removed types of paramaters 2025-06-12 17:14:33 +02:00
1a82d72a43 updated readme 2025-06-12 17:11:05 +02:00
5819458c17 updated readme 2025-06-12 17:09:41 +02:00
72fadaabe8 updated readme 2025-06-12 17:05:06 +02:00
81b2c1f782 updated readme 2025-06-12 16:58:07 +02:00
b0349dc44b updated readme 2025-06-12 16:55:05 +02:00
be2f26a7f6 drafted readme 2025-06-12 16:53:37 +02:00
f1204c4c35 fixed extraction of correct predictions 2025-06-12 13:25:41 +02:00
21de0ffd7a updated logic for the extraction of correct
predictions
2025-06-10 23:41:45 +02:00
f36fcc6e05 updated the way the comment generation and code
refinement inputs are exported (automatized the
putting of archives for context)
2025-06-10 23:40:44 +02:00
f5bdfd1a1b the input to code refinement now ignores paraphrases 2025-06-10 20:45:51 +02:00
429fe9b060 implemented new way to extract stats from dataset 2025-06-10 20:42:58 +02:00
dd52e43000 added way to put paraphrases from external csv 2025-06-10 20:42:55 +02:00
1754f93018 quality of life for manual selection 2025-06-05 10:49:05 +02:00
4c5e486ad6 added small .unique() on repo names to avoid
processing a repo twice
2025-06-05 10:46:21 +02:00
bf1591c61d fixed bug in manual selection 2025-06-04 12:07:41 +02:00
9a24a734e7 added the printing of the relevant hunk when
asking for comment relevance
2025-06-04 10:37:48 +02:00
6110640a6f fixed condition to check whether a comment was
within the diffs
2025-06-03 13:39:02 +02:00
792195e33c now ensuring the comment is within the diff_before
changes
2025-06-03 11:51:38 +02:00
926d3a3681 now using original start line as default and start
line as backup instead of the other way around
2025-06-03 11:51:07 +02:00
154837827d added link to paraphrases extraction 2025-06-03 10:10:57 +02:00
45a8122408 using enum choice actoin instead of the previous
thing we were using
2025-06-03 10:10:36 +02:00
66d046cbaa made filename a positional argument 2025-06-03 10:10:19 +02:00
c05c9cb366 fixed manual selection 2025-06-02 15:33:23 +02:00
87b49b377d the removal of the is_code_related in field in
selection broke backwards compatilibility. Fixed
it
2025-06-02 10:46:47 +02:00
4648ba2560 fixed type annotation 2025-06-02 09:50:09 +02:00
09df9a1ae8 removed already done TODO 2025-06-02 09:49:59 +02:00
5b8357567b removed code relatedness from manual selection
since now it's already done by pull_requests
2025-06-02 09:48:27 +02:00
b311c49f9a simplifying logging of common error we can't do
much about
2025-05-28 10:17:59 +02:00
77ed66ded8 added small logging statement 2025-05-28 10:17:48 +02:00
e097885e36 only writing the dataset to disk when there are new entries 2025-05-28 10:17:19 +02:00
0b182837c1 added new option for dataset 2025-05-27 10:50:10 +02:00
900003bac7 added a way to extract the information to then
generate paraphrases
2025-05-27 10:48:17 +02:00
63b69e40b8 tried to make requests cache better 2025-05-26 11:36:31 +02:00
a4ce620aa0 printing stacktrace when error is made 2025-05-26 11:36:04 +02:00
7e00656ab1 fixed condition 2025-05-26 11:35:54 +02:00
e619d2f339 added another point of failure 2025-05-21 10:59:07 +02:00
5734ca5c8d if we are multithreading, give some time between the requests 2025-05-21 09:40:30 +02:00
b598e97bc6 moved update of pbar 2025-05-21 09:33:44 +02:00
9ced42b6c4 removed unused variable 2025-05-21 09:33:31 +02:00
f1e8b896bb fixed slight bug 2025-05-21 09:18:51 +02:00