How sshema-cpu/heart-disease-ml works
This is an educational heart-disease classification project, centered on a notebook that compares three models using 1,025 patient records. The documented workflow loads the CSV, explores it, splits and scales features, trains models, and evaluates them with test metrics, cross-validation, and visualizations. The report and portfolio page are separate project artifacts. The notebook’s source was not sampled, so workflow links reflect the README’s stated methodology; links to artifact internals are omitted.
Main entry points: heart_disease_prediction.ipynb Split and scale · heart.csv Heart records · Sam_Shema_ML_Final_Report.pdf Written report · index.html Portfolio page
Components 10
Dataset
- Heart records
data/heart.csv
Notebook Analysis
- Explore data
notebooks/heart_disease_prediction.ipynb - Split and scale
notebooks/heart_disease_prediction.ipynb - Logistic regression
notebooks/heart_disease_prediction.ipynb - Decision tree
notebooks/heart_disease_prediction.ipynb - Neural network
notebooks/heart_disease_prediction.ipynb - Evaluate models
notebooks/heart_disease_prediction.ipynb
Project Artifacts
- Written report
report/Sam_Shema_ML_Final_Report.pdf - Portfolio page
index.html
Other components
- Project reader
Connections 11
- Project reader → Explore data: runs notebook
- Heart records → Explore data: loads records
- Explore data → Split and scale: prepares features
- Split and scale → Logistic regression: trains baseline
- Split and scale → Decision tree: trains tree
- Split and scale → Neural network: trains network
- Logistic regression → Evaluate models: evaluates
- Decision tree → Evaluate models: evaluates
- Neural network → Evaluate models: evaluates
- Project reader → Written report: reads
- Project reader → Portfolio page: views