Skip to content

Latest commit

 

History

History
46 lines (35 loc) · 1.58 KB

README.md

File metadata and controls

46 lines (35 loc) · 1.58 KB

Heart Disease Diagnosis

Main Task

  1. Classify patients with heart failure
  2. Identify correlated features

Install

pip install -r requirements.txt

Usage

Train

python main.py -t data/train.csv -v data/valid.csv --lr=0.0002 -b=20 --lr_scheduler cosine --epochs 1000 --suffix hidden_32_16

Evaluate

python main.py -e -v data/test.csv --save_results test_results.csv --lr=0.0002 -b=20 --lr_scheduler cosine --resume models/arch\[NeuralNet\]_optim\[adam\]_lr\[0.0002\]_lrsch\[cosine\]_batch\[20\]_WeightedSampling\[False\]_hidden_32_16/model_best.pth.tar

Code

  1. Data pre-processing [here]
  2. Neural Network [here]
  3. Conventional ML models [here]
  4. Report [here]

Data

  1. processed.cleveland.data file, which is available from the Data Folder
  2. Description file heart-disease.names

Other Published Code

  1. https://github.com/AbdullahAlrhmoun/Heart-disease-prediction-model
  2. https://www.kaggle.com/aavigan/predicting-coronary-heart-disease-non-invasively
  3. https://www.kaggle.com/ronitf/predicting-heart-disease
  4. https://www.kaggle.com/sharansmenon/heart-disease-pytorch-nn
  5. https://github.com/knickhill/heart-disease-classification/blob/master/part2-models.ipynb