GeoSensor-Movement-Predictor

Linear Regression ML Model To Predict Type Of Movement Based On GeoSensor Dataset (walking, running.. )

• Dataset included as CSV
• Model trainer script included
• Model class included
• Predictor script included
• Samples usage included in predictor script

How To Use

Setup

• Intall pyenv: https://github.com/pyenv/pyenv?tab=readme-ov-file#installation

Run the following commands in terminal:

cd ~ && git clone https://github.com/CYFARE/GeoSensor-Movement-Predictor && cd GeoSensor-Movement-Predictor
pyenv install 3.11.11
pyenv local 3.11.11
python -V && python -m venv venv
source venv/bin/activate
python -m pip install -r requirements.txt

• Run: python predictor.py to see model prediction results with sample data provided in the same python script.

Prediction of Live Data

Given types of geosensor data, predict.py can be used to determine at real-time speeds the type of movement - walking, running etc. based on trained dataset. It's trained using linear regression so it's quite accurate.