| This data belong to the EmoPain Challenge 2020 Task 2: Movement Behaviour Classification, Pain Recognition dataset. | |
| https://github.com/Mvrjustid/EmoPainChallenge2020 | |
| Original data constains | |
| • Training set (10 chronic pain participants, 6 healthy participants) | |
| • Validation set (4 chronic pain participants, 3 healthy participants) | |
| Each instance consists of data from 30 sensors measuring features such as elbow and neck angle. | |
| The class values are ordinal and pain levels (Task 2 - Pain Recognition from Movement) | |
| 0: Healthy | |
| 1: Low-level pain | |
| 2: High-level pain. | |
| The original train data are long unequal length series but the final test data is of fixed length 180. We segmented the train and validation data (wiht no overlap) to series of length 180. | |