Open Daivd093 opened 9 months ago
same issue
I checked the type of space
, it is gymnasium.spaces.box.Box. I suppose this should be similar to gym.spaces.Box, so I just modified a little bit to avoid the asserted error.
Hi!
I'm not 100% sure what this issue is. However, note that the code was developed 3+ yrs ago, and so was intended for gym==0.23.1
not gymnasium and development on ubuntu. This error seems to be a result of not having the correct dependencies.
Let me know if that fixes the issue!
I'm trying to reproduce the results of the paper, but as a local job instead of using SLURM. But when I'm trying to do the pretraining it gives me:
Followed by this message:
The input I'm trying to give to the script is as follows:
python tools/run_local.py --entry-point scripts/metaworld/collect_policy_dataset.py --cpus 8 --gpus 1 --valid-gpus 1 --arguments benchmark=ml10 tasks-per-env=25 cross-env-ep=10 within-env-ep=25 expert-ep=15 random-ep=2 epsilon=0.1 num-workers=10 noise-type=gaussian path=datasets/mw
What should I do in order to get the local job to work? Is there something wrong with the input parameters?