diff --git a/README.md b/README.md index 307928cc..172e6e55 100644 --- a/README.md +++ b/README.md @@ -8,7 +8,7 @@

-This library contains a collection of Reinforcement Learning robotic environments that use the [Gymansium](https://gymnasium.farama.org/) API. The environments run with the [MuJoCo](https://mujoco.org/) physics engine and the maintained [mujoco python bindings](https://mujoco.readthedocs.io/en/latest/python.html). +This library contains a collection of Reinforcement Learning robotic environments that use the [Gymnasium](https://gymnasium.farama.org/) API. The environments run with the [MuJoCo](https://mujoco.org/) physics engine and the maintained [mujoco python bindings](https://mujoco.readthedocs.io/en/latest/python.html). The documentation website is at [robotics.farama.org](https://robotics.farama.org/), and we have a public discord server (which we also use to coordinate development work) that you can join here: [https://discord.gg/YymmHrvS](https://discord.gg/YymmHrvS) @@ -41,7 +41,7 @@ The different tasks involve hammering a nail, opening a door, twirling a pen, or ## Multi-goal API -The robotic environments use an extension of the core Gymansium API by inheriting from [GoalEnv](https://robotics.farama.org/content/multi-goal_api/) class. The new API forces the environments to have a dictionary observation space that contains 3 keys: +The robotic environments use an extension of the core Gymnasium API by inheriting from [GoalEnv](https://robotics.farama.org/content/multi-goal_api/) class. The new API forces the environments to have a dictionary observation space that contains 3 keys: * `observation` - The actual observation of the environment * `desired_goal` - The goal that the agent has to achieved diff --git a/docs/content/multi-goal_api.md b/docs/content/multi-goal_api.md index cceeff24..2cae0d6b 100644 --- a/docs/content/multi-goal_api.md +++ b/docs/content/multi-goal_api.md @@ -6,7 +6,7 @@ title: Multi-goal API ## Multi-goal API -The robotic environments use an extension of the core Gymansium API by inheriting from [GoalEnv](https://robotics.farama.org/envs/#) class. The new API forces the environments to have a dictionary observation space that contains 3 keys: +The robotic environments use an extension of the core Gymnasium API by inheriting from [GoalEnv](https://robotics.farama.org/envs/#) class. The new API forces the environments to have a dictionary observation space that contains 3 keys: * `observation` - The actual observation of the environment * `desired_goal` - The goal that the agent has to achieved diff --git a/docs/envs/MaMuJoCo/index.md b/docs/envs/MaMuJoCo/index.md index 4c1e6a55..40ae76c4 100644 --- a/docs/envs/MaMuJoCo/index.md +++ b/docs/envs/MaMuJoCo/index.md @@ -11,9 +11,9 @@ lastpage: MaMuJoCo was introduced in ["FACMAC: Factored Multi-Agent Centralised Policy Gradients"](https://arxiv.org/abs/2003.06709). -There are 2 types of Environments, included (1) multi-agent factorizations of [Gymansium/MuJoCo](https://gymnasium.farama.org/environments/mujoco/) tasks and (2) new complex MuJoCo tasks meant to me solved with multi-agent Algorithms. +There are 2 types of Environments, included (1) multi-agent factorizations of [Gymnasium/MuJoCo](https://gymnasium.farama.org/environments/mujoco/) tasks and (2) new complex MuJoCo tasks meant to me solved with multi-agent Algorithms. -Gymansium-Robotics/MaMuJoCo Represents the first, easy to use Framework for research of agent factorization. +Gymnasium-Robotics/MaMuJoCo Represents the first, easy to use Framework for research of agent factorization. ## API MaMuJoCo mainly uses the [PettingZoo.ParallelAPI](https://pettingzoo.farama.org/api/parallel/), but also supports a few extra functions: diff --git a/docs/index.md b/docs/index.md index 50c667b8..b49ea2f0 100644 --- a/docs/index.md +++ b/docs/index.md @@ -45,7 +45,7 @@ Gymnasium-Robotics is a collection of robotics simulation environments for Reinf :height: 500 ``` -This library contains a collection of Reinforcement Learning robotic environments that use the [Gymansium](https://gymnasium.farama.org/) API. The environments run with the [MuJoCo](https://mujoco.org/) physics engine and the maintained [mujoco python bindings](https://mujoco.readthedocs.io/en/latest/python.html). +This library contains a collection of Reinforcement Learning robotic environments that use the [Gymnasium](https://gymnasium.farama.org/) API. The environments run with the [MuJoCo](https://mujoco.org/) physics engine and the maintained [mujoco python bindings](https://mujoco.readthedocs.io/en/latest/python.html). The creation and interaction with the robotic environments follow the Gymnasium interface: diff --git a/gymnasium_robotics/envs/maze/ant_maze_v5.py b/gymnasium_robotics/envs/maze/ant_maze_v5.py index 1635cee1..ec314c95 100644 --- a/gymnasium_robotics/envs/maze/ant_maze_v5.py +++ b/gymnasium_robotics/envs/maze/ant_maze_v5.py @@ -201,7 +201,7 @@ class AntMazeEnv(MazeEnv, EzPickle): ``` ### Version History - - v5: Is now based on `Gymansium/MuJoCoAnt-v5/`, and inherits all features from it such as the `xml_file` argument for the loading of third party model. + - v5: Is now based on `Gymnasium/MuJoCoAnt-v5/`, and inherits all features from it such as the `xml_file` argument for the loading of third party model. - v4: Refactor compute_terminated in MazeEnv into a pure function compute_terminated and a new function update_goal which resets the goal position. Ant bug fix: Reward is now computed before reset (i.e. sparse reward is not always zero). Maze bug fix: Ant can no longer reset within the goal radius 0.45 due to maze_size_scaling factor missing in MazeEnv. info['success'] key added. - v3: refactor version of the D4RL environment, also create dependency on newest [mujoco python bindings](https://mujoco.readthedocs.io/en/latest/python.html) maintained by the MuJoCo team in Deepmind. - v2 & v1: legacy versions in the [D4RL](https://github.com/Farama-Foundation/D4RL). diff --git a/gymnasium_robotics/envs/multiagent_mujoco/coupled_half_cheetah.py b/gymnasium_robotics/envs/multiagent_mujoco/coupled_half_cheetah.py index 0ab735d3..8174e380 100644 --- a/gymnasium_robotics/envs/multiagent_mujoco/coupled_half_cheetah.py +++ b/gymnasium_robotics/envs/multiagent_mujoco/coupled_half_cheetah.py @@ -146,7 +146,7 @@ def __init__(self, render_mode: typing.Optional[str] = None): """Init. Args: - render_mode: see [Gymansium/MuJoCo](https://gymnasium.farama.org/environments/mujoco/) + render_mode: see [Gymnasium/MuJoCo](https://gymnasium.farama.org/environments/mujoco/) """ self._forward_reward_weight = 1 self._ctrl_cost_weight = 0.1 diff --git a/gymnasium_robotics/envs/multiagent_mujoco/mujoco_multi.py b/gymnasium_robotics/envs/multiagent_mujoco/mujoco_multi.py index 2452de19..b45ef92d 100644 --- a/gymnasium_robotics/envs/multiagent_mujoco/mujoco_multi.py +++ b/gymnasium_robotics/envs/multiagent_mujoco/mujoco_multi.py @@ -9,7 +9,7 @@ - General code cleanup, factorization, type hinting, adding documentation and code comments. - Now uses PettingZoo APIs instead of an original API. - Now supports custom agent factorizations. - - Added `gym_env` argument, which can be used to load third party `Gymansium.MujocoEnv` environments. + - Added `gym_env` argument, which can be used to load third party `Gymnasium.MujocoEnv` environments. This project is covered by the Apache 2.0 License. """ @@ -103,10 +103,10 @@ def __init__( global_categories: The categories of observations extracted from the global observable space, For example: if it is set to `("qpos")` out of the globally observable items of the environment, only the position items will be observed. The default is: `("qpos", "qvel")` - render_mode: See [Gymansium/MuJoCo](https://gymnasium.farama.org/environments/mujoco/), + render_mode: See [Gymnasium/MuJoCo](https://gymnasium.farama.org/environments/mujoco/), valid values: 'human', 'rgb_array', 'depth_array' gym_env: A custom `MujocoEnv` environment, overrides generation of environment by `MaMuJoCo`. - kwargs: Additional arguments passed to the [Gymansium/MuJoCo](https://gymnasium.farama.org/environments/mujoco/) environment, + kwargs: Additional arguments passed to the [Gymnasium/MuJoCo](https://gymnasium.farama.org/environments/mujoco/) environment, Note: arguments that change the observation space will not work. Raises: NotImplementedError: When the scenario is not supported (not part of of the valid values). @@ -197,7 +197,7 @@ def _create_base_gym_env( self, scenario: str, agent_conf: str, render_mode: str, **kwargs ) -> gymnasium.envs.mujoco.mujoco_env.MujocoEnv: """Creates the single agent environments that is to be factorized.""" - # load the underlying single agent Gymansium MuJoCo Environment in `self.single_agent_env` + # load the underlying single agent Gymnasium MuJoCo Environment in `self.single_agent_env` if scenario in _MUJOCO_GYM_ENVIROMENTS: return gymnasium.make(f"{scenario}-v5", **kwargs, render_mode=render_mode) elif scenario in ["ManySegmentAnt"]: