Assessment
Grade Composition
Basic Conversational Agent: 30%
Extensions to the Agent (quality and originality): 30%
Written Report: 30%
Individual Quiz: 10%
A pass adds 1 full point to the grade.
A fail adds no points.
Practical Attendance:
Pass or Fail
Mandatory pass. A failure will lead to failing the entire course.
Midway Check-In Presentation:
Pass or Fail
Mandatory pass. A failure will lead to failing the entire course.
To complete the project, you need to provide the following deliverables; make sure to follow the instructions for providing us with these deliverables at the end of this page:
Instructions
main branch - inclusion, exclusion + extensions
inclusion_only branch
All deliverables should be made available on the main branch of your GitHub repository:
Your final agent code (all the MARBEL source code files including .mas2g, .mod2g, Prolog files, etc.) should be located in the folder named agent in your repository.
...
You should make all the data you collected (testing data, as well as your pilot testing data with users outside your team) available in a folder named data in your repository.
...
Please note your individual Commits will also be checked be sure to read through the page before committing.
Deliverables - Repository
| Note |
|---|
Please put your model checkpoint file on Canvas! If you make changes in Dialogflow export your agent into a zip file and replace the BellaBot zip file in the repository. If you make changes to the repository structure, running process, or install process add a README.md to your repository explaining how to use your agent correctly. If you want us to test or see something specifically add it to the README file. Let us know if you made specific changes in files we should take note of. |
Please ensure the following items are included and properly organized when submitting your project:
...
1. Git Repository
Main Branch:
Contains the full codebase for inclusion, exclusion, and all extensions.
Includes the final pipeline implementation.
with Dialogflow zip file if you made Dialogflow changes.
Branch Structure:
inclusion_only: Contains the codebase focusing solely on inclusion logic (with Dialogflow for the MARBEL agent)inclusion_exclusion: Contains the codebase for inclusion and exclusion logic combined (with Dialogflow or without for the MARBEL agent).
...
2. Report
A detailed report documenting the project’s development and results.
Submit the report in markdown.
...
3. Midway Check-In
Provide the PDF version of your slides from the midway check-in presentation.
...
4. Pipeline Results
Include the output/results from your pipeline to demonstrate functionality and performance.
...
5. Trained Model
Provide the trained model file in Canvas:
Checkpoint File:
model_checkpoint.pt.
...
6. Ontology
If any modifications were made to the ontology, ensure the updated version is included.
...
7. Data
If any changes or additions were made to the dataset, provide the updated or additional data files.
...
Key Notes
Ensure your Git repository is well-organized, with proper branch structure.
Verify that all files mentioned above are present, named appropriately, and accessible.
...
| Note |
|---|
Please put your model checkpoint file on Canvas! If you make changes in Dialogflow export your agent into a zip file and replace the BellaBot zip file in the repository. If you make changes to the repository structure, running process, or install process add a README.md to your repository explaining how to use your agent correctly. If you want us to test or see something specifically add it to the README file. |