brianlmerritt / lmsassistant-api

Repository for the API storing Moodle platform course files in vector storage. It makes it possible to request precise information on the platform's course files using the OpenAI API and, in future, other LLMs.

API Moodle files stored in vector storage

This is a repository for the API storing the files of the courses of Moodle platform in a vector storage. It enables to request accurate information about the files of the courses in the platform thankfully to the OpenAI API.

Installation

Clone the repository and install the requirements:

pip3 install -r requirements.txt

Usage

Set the environnement variables (.env file) :

LLM_PROVIDER=<LLM Provider to use ex: openai>
API_KEY=<API key enabling the access to this API>
MODEL_NAME=<Model name to use ex: gpt-3.5-turbo>
DOC_LANGUAGE=<Language of the documents ex: en>
OPENAI_API_KEY=<OpenAI API key>
MAX_PROMPT_TOKENS=<Max tokens to use in prompt ex: 2048>
MAX_CONCURRENCY=<Max concurrency to use ex: 10>
MAX_QUEUE=<Max queue to use ex: 10>
MODEL_CACHE=<Cache directory to use ex: cache> 
WS_TOKEN=<Moodle Web Service token>
WS_ENDPOINT=<Moodle Web Service endpoint ex: https://moodle.example.com/webservice/rest/server.php>
WS_STORAGE=<Downloading files temporary storage directory ex: moodle_file>

Do not forget to create token in Moodle and give it the right permissions to access to local_lmsassistant function and core_course_get_contents function.

Launch the application:

uvicorn main:app --host 0.0.0.0 --port 8080

Maintainer

Lmsassistant is developed by Pimenko Team. https://pimenko.com/

Support

This project is being developed with the support of Lab e-nov https://www.ifp-school.com/recherche-et-innovation/lab-enov

Any Problems, questions, suggestions

If you have a problem with this theme, suggestions for improvement, drop an email via :

• Github
• Your webiste: https://pimenko.com/