decisionforce/CoPO

Fail to restore data from checkpoint

pengzhenghao opened this issue · 2 comments

pengzhenghao commented

Hi Zhenghao,
I trained the intersection using torch copo (train_copo.py) and tried to evaluate the performance using copo_code/new_vis.py.
However, it gives unpickled = pickle.loads(data) TypeError: an integer is required (got type bytes).
I used the file that was stored in TEST/CoPOTraininger_Multi.../checkpoint_000440/algorithm_state.pkl. Did I use the wrong pkl file?

Originally posted by @XilunZhangRobo in #25 (comment)

pengzhenghao commented

It has been a long time since I wish to refactor the visualization script so the user can restore checkpoint directly using RLLib utility. It's a great opportunity for me to do this now.

@XilunZhangRobo could you please shared your checkpoint folder (in .zip file) here? I will check it.