andreshenao/obesity-weight-regression

This model predicts a person’s weight (kg) from height (m) and age (years) using a Linear Regression model from scikit-learn. It is for educational demonstration in a URI data science course.

Model Performance

The model explains about 28% of weight variation — it recognizes the trend that taller and older people weigh more, but the predictions are not highly accurate.

Evaluation Summary

• —Clear positive correlation in predictions
• —Spread increases for heavier individuals
• —Indicates missing key predictors:
• —Diet
• —Muscle mass
• —Body composition
• —Lifestyle factors like exercise

Strengths & Weaknesses

Strengths

• —Simple and interpretable linear model
• —Works as a learning tool for regression
• —Fast and easy to run

Weaknesses

• —Low predictive power
• —Sensitive to outliers
• —Not reliable for real-world medical prediction

Training Details

• —Model: LinearRegression()
• —Train/Test: 75% / 25%
• —Dataset: ObesityDataSet_raw_and_data_synthetic.csv
• —Frameworks: Python, scikit-learn
• —Random State: 42

⚠ Limitations

• —Should not be used for health decisions
• —Trained on a limited synthetic dataset
• —Does not model non-linear relationships important for weight

How to Use

import skops.io as sio
model = sio.load("regression_model.skops", trusted=True)
prediction = model.predict([[1.75, 21]])
print(prediction)