This repository contains the project for the course Machine learning for medical data of N. Sokolovska at Sorbonne University. More information about the course at https://sites.google.com/view/nsokolovska/teaching?authuser=0
To do this project, we have chosen a classification dataset called "Classify gestures by reading muscle activity" and available at : https://www.kaggle.com/datasets/kyr7plus/emg-4. It aims at doing prediction of gestures (between rock, paper, scissors and ok) based on some reccord of the electrical activity produced by muscles when doing the movements.
- Data for this tasks: data for 4 hand gestures
- metrics: metrics for evaluations
- models: models for making predictions
- preprocessing: data preprocessing
Dependong on the system used, be careful of the path to the data, models and figures. In particular windows are using '/' to separate folders and linux/max ''.