eduNinja is the culmination of years of self-directed and self-funded research, emerging as a comprehensive system for personalized learning support. Focused on a "prioritized assessment" and "offline-first" approach, eduNinja emphasizes STEAM (Science, Technology, Engineering, Arts, and Mathematics). It offers optimized multiple-choice assessments for mathematics and science, integrating an AI-assisted feedback and recommendation system to provide an adaptive and effective learning experience.
eduNinja-Evals facilitates the creation and management of multiple-choice assessments in mathematics and science (STEAM). It allows users to take assessments with instant feedback using a locally deployed machine learning language model via Ollama or Claude 3.5 Sonnet through Anthropic API for AI-assisted feedback.
- Course & Assessment Management: Create and manage courses, questions, and assessments.
- Instant Feedback: Take assessments with immediate AI-assisted feedback.
- AI Integration: Leverage language models for feedback via Ollama or Anthropic API.
- Account Management: Includes functionality for user registration and login.
- LaTeX Rendering: High-quality equation rendering for a better learning experience.
- Question Bank: Access an extensive question bank for assessments.
- Personalized Learning System: Currently a placeholder, with plans for full implementation soon.
eduNinja-PL is a personalized learning support system. Based on user-declared preferences and diagnostic tests, it provides recommendations and feedback from LLMs (Large Language Models).
-
Install Docker.
-
Clone this repository:
git clone https://github.com/sudomakeflies/eduNinja.git
-
Navigate to the project directory:
cd eduNinja -
Copy the example environment file and customize it with your API keys:
mv .env_example .env
-
Build and run the project:
docker-compose up --build
-
Access the application on localhost and start exploring!
Contributions are welcome! If you have ideas for new features, bug fixes, or code improvements, feel free to open an issue or submit a pull request.
This project is licensed under the GPLv3 License.