rskpdev / ECE-143-Project

UCSD ECE Group 19 Project: Prediction and Analysis of Heart Failure

ECE-143-Project

UCSD ECE Group 19 Project: Prediction and Analysis of Heart Failure

Installation

Requires:

• conda
• python 3.9.5
• pandas
• numpy
• jupyter notebook
• sklearn
• seaborn
• plotly

Clone the repository using

git clone https://github.com/rskpdev/ECE-143-Project.git

Create a conda environment from the environment.yml file. The first line of the .yml file sets the new environment's name

conda env create -f environment.yml

Activate the conda environment

conda activate 143_env

Deactivate when done making changes

conda deactivate

Usage

The data we used is stored in data folder, and the Machine Learning model scripts are stored in model folder.

EDA

EDA of the features from dataset are stored in the notebook in notebooks folder.
View notebook here jupyter nbviewer

Machine Learning

Machine Learning of the features from dataset are stored in the notebook in model folder.
View notebook here jupyter nbviewer

• PCA.py
  This file is used for performing PCA to extract the top 2 features for visualization.
  
• model.py
  This file consists of all the models used in prediction, plotting confusion matrix and calculating the metric scores.
  
• encoder.py
  this file encodes the categorical features in dataset.
  
• split_dataset.py
  This file splits the dataset into testing and training sets