The aim of this project is to be able to detect the dance moves by using the data input from gyroscopes and accelerometers attached to both of the user's hands.
Steps: 1: Collecting data, Loading data, preprocessing data, Split data into fixed steps, one-hot encoding 2: Feature extraction.
Training model
- Convolution Neural Networks, Feed Forward Neural Networks Testing model
- K-fold cross validation, classification accuracy, confusion matrix Tuning hyper-parameters
- Window size, overlap size, batch size, layers, nodes, features to use