Diet Recommendation Project v1

This repository contains the codebase for the Diet Recommendation project (v1).

Project Structure

• data/: Directory for all data-related items
  • raw/: Raw data, do not alter, do not upload large files to version control
  • processed/: Processed data, ready for training
  • mappings/: Mappings needed in later processing and training
  • embeddings/: Meal, food, and ingredient embedding and graph data
• models/: Trained model files
• src/: Source files for the project
  • preprocessing/: Scripts for data preprocessing
  • training/: Scripts for model training
  • evaluation/: Scripts for model evaluation
  • utils/: Utility scripts and helper functions
• logs/: Logs from model training and evaluation
• results/: Results from model evaluations

Setup

1. Clone the repository and navigate to the project directory.
2. Ensure you are using Python 3.9.
3. Run pip install -r requirements.txt to install dependencies.
4. Execute bash setup_repo.sh to create the directory structure and initial setup files.

Running the Pipeline

Locally Using Makefile

You can run the entire pipeline or individual stages using the Makefile:

• Run the entire pipeline: make all
• Only preprocess data: make preprocess
• Only train models: make train
• Only evaluate models: make evaluate

Using GitHub Actions

The project is configured with GitHub Actions for Continuous Integration (CI) and Continuous Deployment (CD):

• CI: On every push and pull request to the main branch, the CI pipeline tests the code.
• CD: The CD pipeline can be triggered manually via GitHub Actions to run the full pipeline, or it's scheduled to run automatically at midnight daily.

Data Version Control (DVC)

We use DVC to track data and model versions:

1. Add new data or models: dvc add your_file_or_directory
2. Commit the changes: git add . && git commit -m "Add changes"
3. Push the changes: git push && dvc push

Refer to the DVC documentation for more details.

Experiment Tracking with Weights & Biases (W&B)

This project uses W&B for experiment tracking. Ensure you are logged in to W&B and have it integrated into the training and evaluation scripts.

Refer to the W&B documentation for usage instructions.

Contribution Guidelines

To contribute to this project:

1. Fork the repository and create a new branch (git checkout -b feature/AmazingFeature).
2. Make changes and test them.
3. Commit your changes (git commit -m 'Add some AmazingFeature').
4. Push to the branch (git push origin feature/AmazingFeature).
5. Open a pull request.

License

This project is licensed under the Apache-2.0 License. Please see the LICENSE file for details.