Learning motions from demonstrations and rewards with time-invariant dynamical systems based policies - View it on GitHub