-
Hey, I read your paper thoroughly, but I am still confused about how did you train your supervised agents? Did you just add the teacher forcing based on PPO agents? Or only use teacher forcing to trai…
-
Hello there!
Congratulations on the release . Been using the library for a while and is getting more straight forward and user friendly! 👍 Thanks in advance for the time taken on trackling this chall…
-
Hello i huge problem with offline BC training, that i already mentioned in one of my comments.
Traning with BC offline is very slow 😞
Reaching 1000 setps takes around 5 minutes, while for the same c…
-
AIRL is broken because its inputs are accidentally preprocessed twice by SB helper functions.
-
Hello. About an hour ago I was able to successfully execute a basic hello world example of behavior cloning. However, now that I trying to implement GAIL, I am obtaining the error below. Furthermor…
-
Is the `trainer.gen_algo` the finally trained model on which i can continue training?
I would like to clone a handcrafted controller into a stable-baselines agent and than continue training. It see…
-
There is a large variety of reinforcement libraries using PyTorch, just to name a few
- https://github.com/ray-project/ray
- https://github.com/DLR-RM/stable-baselines3
- https://github.com/ikost…
-
- [ ] I have marked all applicable categories:
+ [ ] exception-raising bug
+ [ ] RL algorithm bug
+ [ ] documentation request (i.e. "X is missing from the documentation.")
+ [ ] ne…
-
on-policy algorithms in this library use a buffer called "RolloutBuffer"
In RL the term rollout is usually used in the context of model-based RL and planning. However, as it is used here its just a…
-
I'm wondering how did you get the data in the "expert" folder?
which RL codebase can provide a demonstration in pkl file?