belinghy/steppingstone

ALLSTEPS: Curriculum-driven Learning of Stepping Stone Skills (SCA 2020)

steppingstone's Issues

Walker3DStepperEnv-v0 train problem

I run Walker3DStepperEnv-v0 train by the following command line/

./scripts/local_run_playground_train.sh curriculum_experiment env_name='mocca_envs:Walker3DStepperEnv-v0' use_curriculum=True

Then, I tried to predict using the following command line.

python3 -m playground.enjoy --env mocca_envs:Walker3DStepperEnv-v0 --net ./runs/2023_03_29__09_00_13__curriculum_experiment/models/mocca_envs:Walker3DStepperEnv-v0_latest.pt

But, the humanoid can not step up the stairs at all, and stopped on the first step.
Please let me know how to train to step up stairs smoothly.

update_specalist not a function of environment MikeStepperEnv-v0.

very excied to use this code to see the demo from https://www.youtube.com/watch?v=Qe1izKvpFRQ using the command:
python3 enjoy.py --env mocca_envs:MikeStepperEnv-v0 --net models/mocca_envs:MikeStepperEnv-v0_latest.pt
but i get an error telling me " mike_StepperEnv object has no attribute 'update_specialist''.

If I delete env.update_specialist() I get further errors related to a missmatch in the size of the environment envs and the expected input size for the nn-model.

help resolving this problem would be much apreciated as I am very excited about trying out this excelent work!!!