jhejna / hierarchical_morphology_transfer Goto Github PK

The paper only contains some of the discriminator hyperparameters

However, there are some other hyperparameters in the README.md, for example

--discrim-decay true
--discrim-online false
--discrim-time-limit 32

So what should I do if I want to reproduce the results in your paper for imitation of AntMaze_Low from PointMaze_Low?
I guess the second row in the table is feasible, right?

I just want to know the command line instructions to reproduce the results in the paper.
I got this error with the command in ReadMe

python scripts/train_model.py --env Ant_High --alg SAC -t 200000 --time-limit 50

gym.error.Error: Attempted to look up malformed environment ID: b'Ant_High'. (Currently all IDs must be of the form ^(?:[\w:-]+\/)?([\w:.-]+)-v(\d+)$.)

I think there is something wrong with this function

I want to render a zero-shot high-level policy after training. In other words, I have trained PointMaze_High and AntMaze_Low policies and want to combine them.

The test_model.py renders only one policy at a time, and the test_hrl_model.py requires the high-level and the low-level environments to match as well.