Wine Chromatic Profile Classification

Authors: Farhan Bin Faisal, Daria Khon, Adrian Leung, Zhiwei Zhang

DSCI 522 (Data Science Workflows) Project

About

In this project we created a classification model to predict the type of wine (red or white) based on its physicochemical properties, such as acidity, sulfates, citric acid, etc. The logistic regression model was selected for its simplicity and effectiveness in binary classification tasks. Our classifier demonstrated excellent performance on unseen data, achieving test accuracy of 0.99. Out of 1,950 instances, only 15 were misclassified, showcasing the model's reliability and its potential for practical applications in wine quality assessment and categorization.

The data set we used in this project was created by By P. Cortez, A. Cerdeira, Fernando Almeida, Telmo Matos, J. Reis. 2009 as part of Decision Support Systems publication, and is available on UCI Machine Learning Repository here.

Report

The final report can be found here.

Dependencies

• Docker
• Jupyter Lab
• or VS CODE with Jupyter Extension

Usage

Note: Docker Desktop is required for this workflow. Make sure it is installed and running before proceeding.

Running the analysis

1. Clone this repository git clone
2. Enter the root project direcory on your local repository and run:

   docker-compose down --rmi all --volumes
   docker-compose up
   

3. In the terminal, look for a URL that starts with http://127.0.0.1:8888/lab?token= (see sample terminal output below). Copy and paste that URL into your browser.

4. Once in a docker container, open a terminal and run:

   make clean
   make all
   

5. To clean up and start fresh at any point, run:

   make clean
   

Clean up

To shut down the container and clean up the resources, type Cntrl + C in the terminal where you launched the container, and then type docker compose rm

Developer notes

Developer dependencies

• conda (version 23.9.0 or higher)
• conda-lock (version 2.5.7 or higher)

Adding a new dependency

1. Add the dependency to the environment.yml file on a new branch.

2. Run conda-lock -k explicit --file environment.yml -p linux-64 to update the conda-linux-64.lock file.

3. Re-build the Docker image locally to ensure it builds and runs properly.

4. Push the changes to GitHub. A new Docker image will be built and pushed to Docker Hub automatically. It will be tagged with the SHA for the commit that changed the file.

5. Update the docker-compose.yml file on your branch to use the new container image (make sure to update the tag specifically).

6. Send a pull request to merge the changes into the main branch.

To run pipeline components individually

4.1 Download, clean, split and validate the data:

   python scripts/download.py \
      --id 186 \
      --save_to ./data/raw/wine.csv

   python scripts/clean_n_split_data.py \
      --raw-data ./data/raw/wine.csv
   
   python scripts/validation_before_split.py \
      --file_name wine.csv \
      --data_path ./data/raw

4.2 Run EDA

   python scripts/eda_n_correlation_check.py \
      --train-file ./data/proc/wine_train.csv \
      --output-img ./results/figures \
      --output-table ./results/tables

4.3 Preprocess data and run it through the Logistic Regression model:

   python scripts/preprocessing.py \
      --pipe-to ./results/models

   python scripts/model_evaluation_wine_predictor.py \
      --train-data ./data/proc/wine_train.csv \
      --test-data ./data/proc/wine_test.csv \
      --pipeline-path ./results/models/wine_pipeline.pickle \
      --table-to ./results/tables \
      --plot-to ./results/figures

4.4 Generate a Quarto report .html and/or .pdf:

   quarto render report/report.qmd --to html
   quarto render report/report.qmd --to pdf

Running the test suite

Use the same docker compose up command as described in the Running the analysis section above to launch Jupyter lab. Tests are run using the pytest command in the tests/ directory. More details about the test suite can be found in tests/.

License

This project was created with the MIT License

References

Cortez P, Cerdeira A, Almeida F, Matos T, Reis J. Wine Quality [dataset]. 2009. UCI Machine Learning Repository. Available from: https://doi.org/10.24432/C56S3T.