Ashvin Nair
Ashvin Nair
I think you just need to smooth (each epoch contains 1 rollout which either succeeds or fails), can you average the returns over a moving window and plot it again?...
You can use https://github.com/vitchyr/viskit or this utility: https://github.com/vitchyr/rlkit/blob/master/rlkit/visualization/plot_util.py to generate plots. The published figures use `plot_util`
Tracked the problem, I messed up and actually didn't realize that catkin_make, leading to the above problem (because the executable never got built, there was no "bundle_calibrate" found). But the...
Sorry for the late response, but the IQL experiments in the paper were run in jax and should be reproducible with the other repo: https://github.com/ikostrikov/implicit_q_learning This reimplementation is pytorch is...