Episode Collection Failure due to mismatch in expected value size in batch.

[MohamedGNabail]

I am trying to reproduce the results for this research:

I am running the local version for collecting the mw dataset:
scripts/metaworld/collect_policy_dataset.py --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=1 --noise-type=gaussian --path=datasets/mw

For the reach-v2 task dataset collection, the collect_episode function raises a value error

ValueError: setting an array element with a sequence. The requested array has an inhomogeneous shape after 1 dimensions. The detected shape was (2,) + inhomogeneous part.

because the value is an array while the expected size for the batch is just an element (start: 0 end end: 1, values passed from the _add_to_buffer function for an int reward)

are any of the steps followed above for reproducing the results wrong ?