Diabetes Prediction Capstone

Predict diabetes using classic ML models on the Pima Indians Diabetes dataset (or a compatible healthcare CSV).

Project Structure

.
├─ README.md
├─ requirements.txt
├─ app.py                      # Streamlit app for inference
├─ .gitignore
├─ LICENSE
├─ diabetes_capstone.ipynb
├─ diabetes_best_model.joblib  # created after running notebook
└─ data/
   └─ README.md               # how to obtain/place datasets

Dataset

• Preferred: Kaggle Pima Indians Diabetes: https://www.kaggle.com/datasets/uciml/pima-indians-diabetes-database
• Or use a compatible CSV placed in Downloads (default search list in the notebook):
  • diabetes.csv
  • Disease_symptom_and_patient_profile_dataset.csv
  • healthcare_dataset.csv

Target column should be one of: Outcome, target, diabetes, or class. Adjust in the notebook if different.

Quickstart

1. Create environment and install dependencies

python3 -m venv .venv
source .venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt

2. Launch the notebook

jupyter notebook diabetes_capstone.ipynb

3. Run all cells

• Performs EDA, preprocessing, trains Logistic Regression / RandomForest / SVM / GradientBoosting
• Compares metrics and saves best model as diabetes_best_model.joblib

4. Export to PDF (for submission)

• From the notebook UI: File → Print Preview → Print to PDF
• Or via CLI:

jupyter nbconvert --to pdf diabetes_capstone.ipynb

Reproducibility

• All steps are captured in the notebook.
• requirements.txt pins core packages for consistent runs.

License

MIT License — see LICENSE.

Streamlit App

Run locally:

pip install -r requirements.txt
streamlit run app.py

Usage:

• Option 1: Upload a CSV with the same feature columns used during training.
• Option 2: Use the manual input form (typical numeric features from Pima dataset).

Deployment (optional):

• Push this repo to GitHub. On Streamlit Cloud (or similar), create a new app pointing to app.py.