-
Notifications
You must be signed in to change notification settings - Fork 5
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
Not an issue... #3
Comments
Sorry to answer that late, I have not recognized you opened this issue, because this has been more of a development repository so far.. |
Really just following up here after the Moodle AI group meeting, and if there's a better place to collaborate please me know. I've installed the code and found it to be really good, so much so I'm inclined to replace the AI provider manager I've written with this! However as I was asking in the video chat, I'm particularly interested in RAG-sort of processes, so I'd be interested in the "purposes" area, and defining a "embeddings" purpose (i.e. using the AI provider to generate a vector representation of the text supplied). |
This place is totally fine :-) Could you elaborate more on how exactly you imagine this to work?
Sorry, I do not have any experience with this kind of use case, but if you can explain in detail what you want to do and how the data flow will be, I'm happy to share how this can be done on the current implementation of the ai manager. |
I've been looking at a mechanism in my AI manager plugin that injects in to Moodle activities some activity settings: These allow the staff member to set conditions for AI use of that particular activity. In this case the "Allow indexing for AI" is intended to be checked by the code that indexes Moodle content, and if it's unchecked then the content of the activity/resource would not be chunked, vectorised or stored in a vector DB. (this has been injected into Moodle Activities in the lib.php. However you can see the other settings here could also be used to influence if the other AI "actions" are permitted by the end-user (student). The Embedding action is also covered by some of the AI APIs already (https://platform.openai.com/docs/guides/embeddings) so the embedding would be passed through to the AI provider class (if it was enabled for that instance). |
Thank you for clarifying. In my head I am now just missing what to do with this, so how does the plugin for the students look like? Is this a chat bot with specific base data or how would this be used? |
But just reaching out to as I've been working on some similar stuff at https://github.com/mhughes2k/moodle-local_ai with respect to Moodle AI management and RAG (https://github.com/mhughes2k/moodle-search_solrrag/) and wonder if there may be some collaboration or conversation ;-)
The text was updated successfully, but these errors were encountered: