Metadata-Version: 2.4
Name: jaxgcrl
Version: 0.2.1
Summary: Blazingly fast goal-conditioned environments based on MJX and BRAX.
License: Apache License
        Version 2.0, January 2004
        http://www.apache.org/licenses/
        
        TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
        
        1. Definitions.
        
        "License" shall mean the terms and conditions for use, reproduction,
        and distribution as defined by Sections 1 through 9 of this document.
        
        "Licensor" shall mean the copyright owner or entity authorized by
        the copyright owner that is granting the License.
        
        "Legal Entity" shall mean the union of the acting entity and all
        other entities that control, are controlled by, or are under common
        control with that entity. For the purposes of this definition,
        "control" means (i) the power, direct or indirect, to cause the
        direction or management of such entity, whether by contract or
        otherwise, or (ii) ownership of fifty percent (50%) or more of the
        outstanding shares, or (iii) beneficial ownership of such entity.
        
        "You" (or "Your") shall mean an individual or Legal Entity
        exercising permissions granted by this License.
        
        "Source" form shall mean the preferred form for making modifications,
        including but not limited to software source code, documentation
        source, and configuration files.
        
        "Object" form shall mean any form resulting from mechanical
        transformation or translation of a Source form, including but
        not limited to compiled object code, generated documentation,
        and conversions to other media types.
        
        "Work" shall mean the work of authorship, whether in Source or
        Object form, made available under the License, as indicated by a
        copyright notice that is included in or attached to the work
        (an example is provided in the Appendix below).
        
        "Derivative Works" shall mean any work, whether in Source or Object
        form, that is based on (or derived from) the Work and for which the
        editorial revisions, annotations, elaborations, or other modifications
        represent, as a whole, an original work of authorship. For the purposes
        of this License, Derivative Works shall not include works that remain
        separable from, or merely link (or bind by name) to the interfaces of,
        the Work and Derivative Works thereof.
        
        "Contribution" shall mean any work of authorship, including
        the original version of the Work and any modifications or additions
        to that Work or Derivative Works thereof, that is intentionally
        submitted to Licensor for inclusion in the Work by the copyright owner
        or by an individual or Legal Entity authorized to submit on behalf of
        the copyright owner. For the purposes of this definition, "submitted"
        means any form of electronic, verbal, or written communication sent
        to the Licensor or its representatives, including but not limited to
        communication on electronic mailing lists, source code control systems,
        and issue tracking systems that are managed by, or on behalf of, the
        Licensor for the purpose of discussing and improving the Work, but
        excluding communication that is conspicuously marked or otherwise
        designated in writing by the copyright owner as "Not a Contribution."
        
        "Contributor" shall mean Licensor and any individual or Legal Entity
        on behalf of whom a Contribution has been received by Licensor and
        subsequently incorporated within the Work.
        
        2. Grant of Copyright License. Subject to the terms and conditions of
        this License, each Contributor hereby grants to You a perpetual,
        worldwide, non-exclusive, no-charge, royalty-free, irrevocable
        copyright license to reproduce, prepare Derivative Works of,
        publicly display, publicly perform, sublicense, and distribute the
        Work and such Derivative Works in Source or Object form.
        
        3. Grant of Patent License. Subject to the terms and conditions of
        this License, each Contributor hereby grants to You a perpetual,
        worldwide, non-exclusive, no-charge, royalty-free, irrevocable
        (except as stated in this section) patent license to make, have made,
        use, offer to sell, sell, import, and otherwise transfer the Work,
        where such license applies only to those patent claims licensable
        by such Contributor that are necessarily infringed by their
        Contribution(s) alone or by combination of their Contribution(s)
        with the Work to which such Contribution(s) was submitted. If You
        institute patent litigation against any entity (including a
        cross-claim or counterclaim in a lawsuit) alleging that the Work
        or a Contribution incorporated within the Work constitutes direct
        or contributory patent infringement, then any patent licenses
        granted to You under this License for that Work shall terminate
        as of the date such litigation is filed.
        
        4. Redistribution. You may reproduce and distribute copies of the
        Work or Derivative Works thereof in any medium, with or without
        modifications, and in Source or Object form, provided that You
        meet the following conditions:
        
        (a) You must give any other recipients of the Work or
        Derivative Works a copy of this License; and
        
        (b) You must cause any modified files to carry prominent notices
        stating that You changed the files; and
        
        (c) You must retain, in the Source form of any Derivative Works
        that You distribute, all copyright, patent, trademark, and
        attribution notices from the Source form of the Work,
        excluding those notices that do not pertain to any part of
        the Derivative Works; and
        
        (d) If the Work includes a "NOTICE" text file as part of its
        distribution, then any Derivative Works that You distribute must
        include a readable copy of the attribution notices contained
        within such NOTICE file, excluding those notices that do not
        pertain to any part of the Derivative Works, in at least one
        of the following places: within a NOTICE text file distributed
        as part of the Derivative Works; within the Source form or
        documentation, if provided along with the Derivative Works; or,
        within a display generated by the Derivative Works, if and
        wherever such third-party notices normally appear. The contents
        of the NOTICE file are for informational purposes only and
        do not modify the License. You may add Your own attribution
        notices within Derivative Works that You distribute, alongside
        or as an addendum to the NOTICE text from the Work, provided
        that such additional attribution notices cannot be construed
        as modifying the License.
        
        You may add Your own copyright statement to Your modifications and
        may provide additional or different license terms and conditions
        for use, reproduction, or distribution of Your modifications, or
        for any such Derivative Works as a whole, provided Your use,
        reproduction, and distribution of the Work otherwise complies with
        the conditions stated in this License.
        
        5. Submission of Contributions. Unless You explicitly state otherwise,
        any Contribution intentionally submitted for inclusion in the Work
        by You to the Licensor shall be under the terms and conditions of
        this License, without any additional terms or conditions.
        Notwithstanding the above, nothing herein shall supersede or modify
        the terms of any separate license agreement you may have executed
        with Licensor regarding such Contributions.
        
        6. Trademarks. This License does not grant permission to use the trade
        names, trademarks, service marks, or product names of the Licensor,
        except as required for reasonable and customary use in describing the
        origin of the Work and reproducing the content of the NOTICE file.
        
        7. Disclaimer of Warranty. Unless required by applicable law or
        agreed to in writing, Licensor provides the Work (and each
        Contributor provides its Contributions) on an "AS IS" BASIS,
        WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
        implied, including, without limitation, any warranties or conditions
        of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
        PARTICULAR PURPOSE. You are solely responsible for determining the
        appropriateness of using or redistributing the Work and assume any
        risks associated with Your exercise of permissions under this License.
        
        8. Limitation of Liability. In no event and under no legal theory,
        whether in tort (including negligence), contract, or otherwise,
        unless required by applicable law (such as deliberate and grossly
        negligent acts) or agreed to in writing, shall any Contributor be
        liable to You for damages, including any direct, indirect, special,
        incidental, or consequential damages of any character arising as a
        result of this License or out of the use or inability to use the
        Work (including but not limited to damages for loss of goodwill,
        work stoppage, computer failure or malfunction, or any and all
        other commercial damages or losses), even if such Contributor
        has been advised of the possibility of such damages.
        
        9. Accepting Warranty or Additional Liability. While redistributing
        the Work or Derivative Works thereof, You may choose to offer,
        and charge a fee for, acceptance of support, warranty, indemnity,
        or other liability obligations and/or rights consistent with this
        License. However, in accepting such obligations, You may act only
        on Your own behalf and on Your sole responsibility, not on behalf
        of any other Contributor, and only if You agree to indemnify,
        defend, and hold each Contributor harmless for any liability
        incurred by, or claims asserted against, such Contributor by reason
        of your accepting any such warranty or additional liability.
        
        END OF TERMS AND CONDITIONS
        
        APPENDIX: How to apply the Apache License to your work.
        
        To apply the Apache License to your work, attach the following
        boilerplate notice, with the fields enclosed by brackets "[]"
        replaced with your own identifying information. (Don't include
        the brackets!)  The text should be enclosed in the appropriate
        comment syntax for the file format. We also recommend that a
        file or class name and description of purpose be included on the
        same "printed page" as the copyright notice for easier
        identification within third-party archives.
        
        Copyright 2023 FLAIR
        
        Licensed under the Apache License, Version 2.0 (the "License");
        you may not use this file except in compliance with the License.
        You may obtain a copy of the License at
        
        http://www.apache.org/licenses/LICENSE-2.0
        
        Unless required by applicable law or agreed to in writing, software
        distributed under the License is distributed on an "AS IS" BASIS,
        WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
        See the License for the specific language governing permissions and
        limitations under the License.
Requires-Python: <3.11,>=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: jax==0.4.25
Requires-Dist: flax>=0.8.3
Requires-Dist: brax>=0.12.1
Requires-Dist: mujoco==3.2.7
Requires-Dist: mujoco-mjx==3.2.7
Requires-Dist: scipy>=1.12.0
Requires-Dist: matplotlib==3.8.4
Requires-Dist: tyro>=0.9.16
Requires-Dist: wandb
Requires-Dist: wandb_osh
Dynamic: license-file

<h1 align="center"> JaxGCRL</h1>


<p align="center">
    <a href= "https://arxiv.org/abs/2408.11052">
        <img src="https://img.shields.io/badge/arXiv-2311.10090-b31b1b.svg" /></a>
    <a href= "https://github.com/MichalBortkiewicz/JaxGCRL/blob/master/LICENSE">
        <img src="https://img.shields.io/badge/license-Apache2.0-blue.svg" /></a>
    <a href= "https://michalbortkiewicz.github.io/JaxGCRL/">
        <img src="https://img.shields.io/badge/docs-green" /></a>
    <a href= "https://michalbortkiewicz.github.io/JaxGCRL/">
        <img src="https://img.shields.io/badge/website-purple" /></a>
</p>


<p align="center"><img src="imgs/grid_transparent.png" width=85%></p>

<center>

[**Installation**](#Installation) | [**Quick Start**](#start) | [**Environments**](#envs) | [**Baselines**](#baselines) | [**Citation**](#cite)

</center>

<br/>

## Accelerating Goal-Conditioned RL Algorithms and Research

We provide blazingly fast goal-conditioned environments based on [MJX](https://mujoco.readthedocs.io/en/stable/mjx.html) and [BRAX](https://github.com/google/brax) for 
quick experimentation with goal-conditioned self-supervised reinforcement learning.

- **Blazingly Fast Training** - Train 10 million environment steps in 10 
  minutes on a single GPU, up to $22\times$ faster than prior implementations.
- **Comprehensive Benchmarking** - Includes 10+ diverse environments and multiple pre-implemented baselines for out-of-the-box evaluation.
- **Modular Implementation** - Designed for clarity and scalability, 
  allowing for easy modification of algorithms.


## Installation 📂
The environment can be set up from the `environment.yml` file.
```bash
conda env create -f environment.yml
```

<h3 name="start" id="start">Quick Start 🚀 </h3>

To verify the installation, run a test experiment using the `./scripts/train.sh` file:

```bash
chmod +x ./scripts/train.sh; ./scripts/train.sh
```
> [!NOTE]  
> If you haven't yet configured [`wandb`](https://wandb.ai/site), you may be prompted to log in.

Specific configs can be specified as in `scripts/train.sh`. The descriptions of the available flags are in `utils.py:create_parser()`. Common flags you may want to change include:
- **env=...**: replace "ant" with any environment name. See `utils.py:create_env()` for names.
- Removing **--log_wandb**: omits logging, if you don't want to use a wandb account.
- **--total_env_steps**: shorter or longer runs.
- **--num_envs**: based on how many environments your GPU memory allows.
- **--contrastive_loss_fn, --energy_fn, --h_dim, --n_hidden, etc.**: algorithmic and architectural changes.

### Environment Interaction

Environments can be controlled with the `reset` and `step` functions. These methods return a state object, which is a dataclass containing the following fields:

`state.pipeline_state`: current, internal state of the environment\
`state.obs`: current observation\
`state.done`: flag indicating if the agent reached the goal\
`state.metrics`: agent performance metrics\
`state.info`: additional info

The following code demonstrates how to interact with the environment:

```python
import jax
from utils.env import create_env

key = jax.random.PRNGKey(0)

# Initialize the environment
env = create_env('ant')

# Use JIT compilation to make environment's reset and step functions execute faster
jit_env_reset = jax.jit(env.reset)
jit_env_step = jax.jit(env.step)

NUM_STEPS = 1000

# Reset the environment and obtain the initial state
state = jit_env_reset(key)

# Simulate the environment for a fixed number of steps
for _ in range(NUM_STEPS):
    # Generate a random action
    key, key_act = jax.random.split(key, 2)
    random_action = jax.random.uniform(key_act, shape=(8,), minval=-1, maxval=1)
    
    # Perform an environment step with the generated action
    state = jit_env_step(state, random_action)
```

### Wandb support 📈
We strongly recommend using Wandb for tracking and visualizing results ([Wandb support](##wandb-support)). Enable Wandb logging with the `--log_wandb` flag. The following flags are also available to organize experiments:
- `--project_name`
- `--group_name`
- `--exp_name`

The `--log_wandb` flag logs metrics to Wandb. By default, metrics are logged to a CSV.

1. Run example [`sweep`](https://docs.wandb.ai/guides/sweeps):
```bash
wandb sweep --project example_sweep ./scripts/sweep.yml
```
2. Then run `wandb agent` with :
```
wandb agent <previous_command_output>
```

We also render videos of the learned policies as `wandb` artifacts. 

<p align="center">
  <img src="imgs/wandb.png" width=55% />
  <img src="imgs/push.gif" width=40%  /> 
</p>

<h2 name="envs" id="envs">Environments 🌎</h2>

We currently support a variety of continuous control environments:
- Locomotion: Half-Cheetah, Ant, Humanoid
- Locomotion + task: AntMaze, AntBall (AntSoccer), AntPush, HumanoidMaze
- Simple arm: Reacher, Pusher, Pusher 2-object
- Manipulation: Reach, Grasp, Push (easy/hard), Binpick (easy/hard)


| Environment     | Env name                                                                 | Code                                              |
| :-------------- | :----------------------------------------------------------------------: | :-----------------------------------------------: |
| Reacher         | `reacher`                                                                | [link](./jaxgcrl/envs/reacher.py)                         |
| Half Cheetah    | `cheetah`                                                                | [link](./jaxgcrl/envs/half_cheetah.py)                    |
| Pusher          | `pusher_easy` <br> `pusher_hard`                                         | [link](./jaxgcrl/envs/pusher.py)                          |
| Ant             | `ant`                                                                    | [link](./jaxgcrl/envs/ant.py)                             |
| Ant Maze        | `ant_u_maze` <br> `ant_big_maze` <br> `ant_hardest_maze`                 | [link](./jaxgcrl/envs/ant_maze.py)                        |
| Ant Soccer      | `ant_ball`                                                               | [link](./jaxgcrl/envs/ant_ball.py)                        |
| Ant Push        | `ant_push`                                                               | [link](./jaxgcrl/envs/ant_push.py)                        |
| Humanoid        | `humanoid`                                                               | [link](./jaxgcrl/envs/humanoid.py)                        |
| Humanoid Maze   | `humanoid_u_maze` <br> `humanoid_big_maze` <br>`humanoid_hardest_maze`   | [link](./jaxgcrl/envs/humanoid_maze.py)                   |
| Arm Reach       | `arm_reach`                                                              | [link](./jaxgcrl/envs/manipulation/arm_reach.py)          |
| Arm Grasp       | `arm_grasp`                                                              | [link](./jaxgcrl/envs/manipulation/arm_grasp.py)          |
| Arm Push        | `arm_push_easy` <br> `arm_push_hard`                                     | [link](./jaxgcrl/envs/manipulation/arm_push_easy.py)      |
| Arm Binpick     | `arm_binpick_easy` <br> `arm_binpick_hard`                               | [link](./jaxgcrl/envs/manipulation/arm_binpick_easy.py)   |

To add new environments: add an XML to `envs/assets`, add a python environment file in `envs`, and register the environment name in `utils.py`.

<h2 name="baselines" id="baselines">Baselines 🤖</h2>

We currently support following algorithms:

| Algorithm                                       | How to run                               | Code                                       |
| ----------------------------------------------- | ---------------------------------------- | ------------------------------------------ |
| [CRL](https://arxiv.org/abs/2206.07568)         | `python run.py crl ...`                 | [link](./jaxgcrl/agents/crl/)                      |
| [PPO](https://arxiv.org/abs/1707.06347)         | `python run.py ppo ...`                 | [link](./jaxgcrl/agents/ppo/)                      |
| [SAC](https://arxiv.org/abs/1801.01290)         | `python run.py sac ...`                 | [link](./jaxgcrl/agents/sac/)                      |
| [SAC + HER](https://arxiv.org/abs/1707.01495)   | `python run.py sac ... --use_her`       | [link](./jaxgcrl/agents/sac/)                      |
| [TD3](https://arxiv.org/pdf/1802.09477)         | `python run.py td3 ...`                 | [link](./jaxgcrl/agents/td3/)                      |
| [TD3 + HER](https://arxiv.org/abs/1707.01495)   | `python run.py td3 ... --use_her`       | [link](./jaxgcrl/agents/td3/)                      |


## Code Structure 📝

The core structure of the codebase is as follows:

<pre><code>
<b>run.py:</b> Takes the name of an agent and runs with the specified configs.
<b>agents/</b>
├── <b>agents/</b>
│   ├── <b>crl/</b> 
│   │   ├── <b>crl.py</b> CRL algorithm 
│   │   ├── <b>losses.py</b> contrastive losses and energy functions
│   │   └── <b>networks.py</b> CRL network architectures
│   ├── <b>ppo/</b> 
│   │   └── <b>ppo.py</b> PPO algorithm 
│   ├── <b>sac/</b> 
│   │   ├── <b>sac.py</b> SAC algorithm
│   │   └── <b>networks.py</b> SAC network architectures
│   └── <b>td3/</b> 
│       ├── <b>td3.py</b> TD3 algorithm
│       ├── <b>losses.py</b> TD3 loss functions
│       └── <b>networks.py</b> TD3 network architectures
├── <b>utils/</b>
│   ├── <b>config.py</b> Base run configs
│   ├── <b>env.py</b> Logic for rendering and environment initialization
│   ├── <b>replay_buffer.py:</b> Contains replay buffer, including logic for state, action, and goal sampling for training.
│   └── <b>evaluator.py:</b> Runs evaluation and collects metrics.
├── <b>envs/</b>
│   ├── <b>ant.py, humanoid.py, ...:</b> Most environments are here.
│   ├── <b>assets:</b> Contains XMLs for environments.
│   └── <b>manipulation:</b> Contains all manipulation environments.
└── <b>scripts/train.sh:</b> Modify to choose environment and hyperparameters.
</code></pre>

The architecture can be adjusted in `networks.py`.


## Contributing 🏗️
Help us build JaxGCRL into the best possible tool for the GCRL community.
Reach out and start contributing or just add an Issue/PR!

- [x] Add Franka robot arm environments. [Done by SimpleGeometry]
- [x] Get around 70% success rate on Ant Big Maze task. [Done by RajGhugare19]
- [ ] Add more complex versions of Ant Sokoban.
- [ ] Integrate environments: 
    - [ ] Overcooked 
    - [ ] Hanabi
    - [ ] Rubik's cube
    - [ ] Sokoban

To run tests (make sure you have access to a GPU):
```bash
python -m pytest 
```

<h2 name="cite" id="cite">Citing JaxGCRL 📜 </h2>
If you use JaxGCRL in your work, please cite us as follows:

```bibtex
@inproceedings{bortkiewicz2025accelerating,
    author    = {Bortkiewicz, Micha\l{} and Pa\l{}ucki, W\l{}adek and Myers, Vivek and
                 Dziarmaga, Tadeusz and Arczewski, Tomasz and Kuci\'{n}ski, \L{}ukasz and
                 Eysenbach, Benjamin},
    booktitle = {{International Conference} on {Learning Representations}},
    title     = {{Accelerating Goal-Conditioned RL Algorithms} and {Research}},
    url       = {https://arxiv.org/pdf/2408.11052},
    year      = {2025},
}
```

## Questions ❓
If you have any questions, comments, or suggestions, please reach out to Michał Bortkiewicz ([michalbortkiewicz8@gmail.com](michalbortkiewicz8@gmail.com)).


## See Also 🙌
There are a number of other libraries which inspired this work, we encourage you to take a look!

JAX-native algorithms:
- [Mava](https://github.com/instadeepai/Mava): JAX implementations of IPPO and MAPPO, two popular MARL algorithms.
- [PureJaxRL](https://github.com/luchris429/purejaxrl): JAX implementation of PPO, and demonstration of end-to-end JAX-based RL training.
- [Minimax](https://github.com/facebookresearch/minimax/): JAX implementations of autocurricula baselines for RL.
- [JaxIRL](https://github.com/FLAIROx/jaxirl?tab=readme-ov-file): JAX implementation of algorithms for inverse reinforcement learning.

JAX-native environments:
- [Gymnax](https://github.com/RobertTLange/gymnax): Implementations of classic RL tasks including classic control, bsuite and MinAtar.
- [Jumanji](https://github.com/instadeepai/jumanji): A diverse set of environments ranging from simple games to NP-hard combinatorial problems.
- [Pgx](https://github.com/sotetsuk/pgx): JAX implementations of classic board games, such as Chess, Go and Shogi.
- [Brax](https://github.com/google/brax): A fully differentiable physics engine written in JAX, features continuous control tasks.
- [XLand-MiniGrid](https://github.com/corl-team/xland-minigrid): Meta-RL gridworld environments inspired by XLand and MiniGrid.
- [Craftax](https://github.com/MichaelTMatthews/Craftax): (Crafter + NetHack) in JAX.
- [JaxMARL](https://github.com/FLAIROx/JaxMARL): Multi-agent RL in Jax.

