-
Notifications
You must be signed in to change notification settings - Fork 7
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Teacher Regrading #12
Comments
Thank you for your feedback. Can you confirm which branch you are working with. I am about to move a large about of updates from this branch https://github.com/marcusgreen/moodle-qtype_aitext/tree/405_core_ai into main. That code has been worked on by another organisation with a large number of users so it may be more robust. |
I've been working on the main branch. |
OK, well I will be incorporating the big set of changes into main within the next two weeks (probably) I hope to keep what is in the main branch as a new branch called legacy. I'm not in a position to look into the bug you reported as moving to the new version will be very time consuming. However I can give you access to an installation with the new branch as a teacher and you can experiment with that if it would help. |
That would be great. |
Hi There Again, I have solved the issue with AI-generated feedback ( Initially, -comment was being manually inserted into the To address this, I created a new database table named |
Thanks for the feedback. |
I have now switched the main branch to contain a lot of code from mebis-lp. I have not tried regrading by hand with this. I suggest upgrading to the lastest version of master and let me know if it works as you expect. Best wishes Marcus |
Thank you Marcus for your update. I've been developing the legacy branch for a while for our own needs. I will implement the new changes into my code. |
Hi,
I believe I've encountered a bug. When a teacher regrades a student's attempt, it updates the marks correctly new from the AI response but doesn't save the new feedback comment generated by the AI. However, performing a dry run (which processes everything from scratch) works as expected.
While debugging the grade_response function, I noticed that the array was null when regrading without a dry run. Since we're manually adding comments to question_attempt_step_data, it seems the insertion process might be where the issue lies. I couldn't fully figure it out.
The text was updated successfully, but these errors were encountered: