Follow up instructions https://apple.github.io/turicreate/docs/userguide/activity_classifier/
Matrix(N*M) where N = E * T where E: number of samples of both drift and no drift and T: the time window. M: {acc_x, acc_y, acc_z, rotation_x, rotation_y, rotation_z, exp_id, activity}
HAPT is the sample provided by apple at the following link https://apple.github.io/turicreate/docs/userguide/activity_classifier/
Use the Jupyter notebook "Turicreate Drift" which has further information of the data
contains the sample data
The testing CoreML model.
The swift which utilizes both Core Motion and CoreML