This application is designed to automate the grading process of Canvas discussions. It uses advanced AI models to grade student responses based on a provided rubric, and also provides developmental feedback to students. The application is built using the Gradio SDK.
- Automated Grading: The application grades Canvas discussions based on a provided rubric, saving educators valuable time.
- Developmental Feedback: In addition to grading, the application provides feedback to students, helping them understand how they can improve.
- Easy to Use: Simply provide the URL of the Canvas discussion and your Canvas API Key, and the application will handle the rest.
-
Ingest Canvas Discussions: Enter your Canvas Discussion URL and your Canvas API Key, then click the 'Ingest' button. The application will download the discussion data and prepare it for grading.
-
Start Grading: Once the data has been ingested, click the 'Grade' button to start the grading process. The application will grade each student's response based on the provided rubric.
-
Download Results: After grading is complete, you can download the results as a CSV file. This file contains the grades for each student, along with the feedback provided by the application.
-
Ask Questions: You can ask questions about the grading process or the results. The application uses an AI model to provide answers to your questions.
This application requires the following Python packages:
- gradio
- openai
- tiktoken
- faiss-cpu
- bs4
- pathvalidate
- unstructured
You can install these packages using pip:
pip install -r requirements.txtTo run the application, simply execute the app.py script:
python app.pyThis will launch the Gradio interface, where you can interact with the application on http://localhost:7860.
Contributions are welcome! If you have any suggestions or improvements, feel free to open an issue or submit a pull request.
This project is licensed under the MIT License.