Skip to content

Latest commit

 

History

History
50 lines (40 loc) · 1.23 KB

README.md

File metadata and controls

50 lines (40 loc) · 1.23 KB

Graphene_BP

The main file to run is called BP_BioZ_v0.py

Installation of Outside Packages

Use the package manager pip to install packages.

pip install numpy 
pip install pandas
pip install hjson
pip install scikit_learn
pip install tensorflow
pip install xgboost

Requirements

Extracted through pypreqs

xgboost==1.4.2
numpy==1.20.3
pandas==1.2.4
hjson==3.0.2
tensorflow==2.3.0
scikit_learn==0.24.2

Configuration

There are several parameters to be configured before running the BP estimation from Bio-Z. If you want to regenerate the original results you can keep the training-related parameters untouched.

To configure the parameters run:

python configure.py

This will walk you through setting the parameters. To use default values for each parameter, showed in the brackets, press "Enter" without entering any character.

example:

Use average features instead of beat-to-beat? (true/false)
features_mean [true]:

Usage

To run BP estimation from Bio-Z:

python BP_BioZ_v0.py

The results including the BP estimation for each subject and the overal error metrics will be stored in folder: './Data/predictions/'