CSC310-fall25/wine-quality-regression

Model description

This is a Linear Regression model trained on combined red and white wine quality data from UCI Machine Learning Repo. The goal of this model is to predict wine quality scores (3-9) based on 12 physicochemical features including wine type.

Intended uses & limitations

[More Information Needed]

Training Procedure

[More Information Needed]

Hyperparameters

Click to expand

Hyperparameter	Value
copy_X	True
fit_intercept	True
n_jobs	None
positive	False
tol	1e-06

Model Plot

LinearRegression()

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

Evaluation Results

[More Information Needed]

How to Get Started with the Model

Start by making a notebook for your eval, then use this starter code:

from huggingface_hub import hf_hub_download
import skops.io as sio
import pandas as pd

# Download model and test data
hf_hub_download(repo_id='CSC310-fall25/wine-quality-regression', filename='model.pkl', local_dir='.')
hf_hub_download(repo_id='CSC310-fall25/wine-quality-regression', filename='test_data.csv', local_dir='.')

# Load model and data
model = sio.load('model.pkl')
test_data = pd.read_csv('test_data.csv')

# Prepare features and target
X_test = test_data.drop('quality', axis=1)
y_test = test_data['quality']

# Make predictions
y_pred = model.predict(X_test)

Model Card Authors

Christian Romualdo

Model Card Contact

cromualdo@uri.edu

Citation

This dataset is from UCI Machine Learning Repository. To learn more, visit: https://archive.ics.uci.edu/dataset/186/wine+quality

Intended uses & limitations

This model is made for educational purposes and is not ready to be used in production.

Training Procedure

I used the scikit-learn linear regression model on a dataset of 5,320 wine samples. The data was split into 80% training and 20% testing, with the training set further split into 75%/25% for validation. The target value is quality and there are 12 features (11 numeric + 1 categorical for wine type). Evaluation metrics used are MAE and R² score.

Evaluation Results

The model achieved an R² score of 0.284 and MAE of 0.580 points on the validation set. The R² score indicates that the model explains about 28.4% of the variance in wine quality. While the model captures some relationships between physicochemical properties and quality, the moderate performance suggests that linear regression may be too simple for this complex task. The residual plots show patterns indicating that a more complex model might better capture the underlying relationships.