Metaworld
Metaworld copied to clipboard
Farama-Foundation
Collections of robotics environments geared towards benchmarking multi-task and meta reinforcement learning
See https://github.com/rlworkgroup/metaworld/pull/154#discussion_r459213695
The issue would be complete when metaworld can pass the `.pre-commit-config.yaml` from https://github.com/rlworkgroup/garage
Currently, the benchmark tasks on support state observations / goals. However, a lot of RL researchers would benefit from the ability to get image observations / goals. There's an ImageEnv...
The only public members of envs should be `step()`, `reset()`, `render()`, `close()`, `action_space`, `observation_space`, and `set_task()`
Hello, Thank you for the excellent code. I'm trying to use the setup to collect simulated RGB data for a project but I haven't been able to find a way...
Hi. I've been trying to reproduce the results of meta-world paper with RLLib. (Figure 11) in [this paper](https://arxiv.org/pdf/1910.10897.pdf) When sometimes during the training procedure, the value error appears. The error...
I think this would be really cool, and good to track the progress of the community. A possible service that already exists for hosting this kind of thing: https://eval.ai/web/challenges/list
I attempted to move the task creation mechanism from MW into each class and these extra classes are a leftover artifact of that.
code: from metaworld.policies.sawyer_stick_push_policy import SawyerStickPushV1Policy ModuleNotFoundError: No module named 'metaworld.policies.sawyer_stick_push_policy' How can I solve this problem? My metaworld version is 2.0.0.
Metadata
Owner
Metadata
Collections of robotics environments geared towards benchmarking multi-task and meta reinforcement learning