CODES Dataset: lotka_volterra

Toy dataset for the CODES benchmark (https://github.com/robin-janssen/CODES-Benchmark). Lotka-Volterra predator-prey model with three predators (p1, p2, p3) and three prey (q1, q2, q3). The system is defined by its gradients as follows: [dp1/dt, dp2/dt, dp3/dt, dq1/dt, dq2/dt, dq3/dt] = [ 0.5 * p1 - 0.02 * p1 * q1 - 0.01 * p1 * q2, 0.6 * p2 - 0.03 * p2 * q1 - 0.015 * p2 * q3, 0.4 * p3 - 0.01 * p3 * q2 - 0.03 * p3 * q3, -0.1 * q1 + 0.005 * p1 * q1 + 0.007 * p2 * q1, -0.08 * q2 + 0.006 * p1 * q2 + 0.009 * p3 * q2, -0.12 * q3 + 0.008 * p2 * q3 + 0.01 * p3 * q3, ] 101 timesteps in the interval [0, 100], solved with scipy.integrate.solve_ivp with the DOP853 method (atol = rtol = 1e-6). Initial conditions for each species sobol-sampled in log-space between 1 and 10. 700/100/200 train/test/val samples.