Skip to content

COSC 4337: DS 2 XGBoost Machine Learning Model + FastAPI

Notifications You must be signed in to change notification settings

BvChung/heart_health

Repository files navigation

Heart Failure Patient Classification Machine Learning Model

Project Overview

Exploratory Data Analysis (EDA) and Machine Learning (ML) models were utilized to predict the death of heart failure patients given this medical record dataset. The dataset was initially preprocessed eliminating missing values and outliers. The ML models used were Logistic Regression, K Nearest Neighbors, Random Forest, and XGBoost. The preprocessed dataset was split into training/validation and test subsets at a ratio of 70:30. Hyperparameter tuning with cross validation was performed and performance metrics were evaluated on the test subset containing unseen data. The best model for this classification problem balancing model complexity in terms of bias and variance was XGBoost.

Code Structure

.
├── heart_failure_eda.ipynb
├── heart_failure_ml_models.ipynb
├── ml_api
│   ├── main.py
│   ├── requirements.txt
│   └── xgb_ml_model.json
└── preprocessed_heart_failure_dataset.csv

About

COSC 4337: DS 2 XGBoost Machine Learning Model + FastAPI

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages