YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
10-Year Heart Disease Risk (Framingham) Case Study
This repository contains a collection of 5 trained models comparing traditional Machine Learning algorithms against Deep Learning variants on an imbalanced tabular healthcare dataset.
Dataset Profile
- Task: Binary Classification (
TenYearCHD) - Rows: 4,238
- Features: 15 clinical/demographic columns
Performance Evaluation Summary
The models were tracked across Accuracy and F1-Score:
- Logistic Regression (Baseline): Accuracy 0.718 | F1-Score 0.055
- Random Forest (Baseline): Accuracy 0.703 | F1-Score 0.067
- Random Forest (Tuned): Accuracy 0.700 | F1-Score 0.073
- MLP/ANN (Baseline): Accuracy 0.683 | F1-Score 0.220
- MLP/ANN (Tuned): Accuracy 0.705 | F1-Score 0.023
Technical Takeaway
Due to class imbalance (~15% positive cases), standard accuracy measures are deceptive. Baseline MLP achieved the highest F1-Score optimization prior to hyperparameter tightening.
- Downloads last month
- 2
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support