Repository to accompany the thesis: Understanding Collective Behvaiour Through Machine Learning.
pip install git+https://github.com/crimbs/kernlearn.git
The dataset can be downloaded from the ScholarsArchive@OSU here.
The data is a JSON format file containing the position, velocity, and fish identifier data for 300 golden shiners in a shallow (depth of 4.5 to 5 cm) rectangular water tank (2.1 by 1.2 meters). There are 5000 individual frames (samples of position and velocity) corresponding to video taken at a rate of 30 frames/s and analysed to extract individual fish’s trajectories. The fields px, py, vx, vy correspond to the x- and y-components of each detected fish’s position (at center of mass) and velocity. The onfish field gives each fish in the frame a unique identifier. When an individual can no longer be distinguished, its identifier is retired; once the fish is again being tracked it is assigned a new ID. This data is a subsample of the frames acquired by the creators in the study https://doi.org/10.1073/pnas.1107583108, refer to it and its supplementary information for more details.