https://github.com/vikashplus/mabel

Manipulation Benchmarks for Learning
https://github.com/vikashplus/mabel

Last synced: about 2 months ago
JSON representation

Manipulation Benchmarks for Learning

• Host: GitHub
• URL: https://github.com/vikashplus/mabel
• Owner: vikashplus
• License: apache-2.0
• Created: 2020-04-14T00:01:03.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2020-06-11T14:55:30.000Z (almost 5 years ago)
• Last Synced: 2025-03-23T18:06:15.731Z (2 months ago)
• Language: Python
• Size: 1.82 MB
• Stars: 4
• Watchers: 5
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# MABEL

MAnipulation BEnchmarks for Learning

`MABEL` is a collection of environments/tasks for benchmarking manipulation, simulated with the [Mujoco](http://www.mujoco.org/) physics engine and wrapped in the [OpenAI `gym`](https://gym.openai.com/) API. 'MABEL' constains following tasks

Pointing           | Pouring       | Relocation             | Zipping

:-------------------------:|:-------------------------:|:-------------------------:|:-------------------------:

![Alt text](mabel/mabel_envs/reach/reach.png?raw=false "Fetch Reaching") |  ![Alt text](mabel/mabel_envs/pouring/pouring.png?raw=false "Liquid pouring") | ![Alt text](mabel/mabel_envs/relocate/relocate.png?raw=false "Object relocation") | ![Alt text](mabel/mabel_envs/zipper/zipper.png?raw=false "Fetch zipper")

## Getting Started

`MABEL` uses git submodules to resolve dependencies. Please follow steps exactly as below to install correctly.

0. Ensure you have access these repositories - [Franka_sim](https://github.com/vikashplus/franka_sim), [Furniture_sim](https://github.com/vikashplus/furniture_sim) and [Scene_sim](https://github.com/vikashplus/scene_sim)

1. Clone this repo with pre-populated submodule dependencies

```

$ git clone --recursive [email protected]:vikashplus/mabel.git

```

2. Update submodules

```

$ cd mabel  

$ git submodule update --remote

```

3. Add repo to pythonpath by updating `~/.bashrc` or `~/.bash_profile`

```

export PYTHONPATH=":$PYTHONPATH"

```

4. Follow install instructions from [mjrl](https://github.com/aravindr93/mjrl) to get model free agents for `MABEL`

5. To visualize an env using a random policy

```

MJPL python mabel/mabel_agents/mjrl/examine_policy.py -i mabel -e MableReachRandom-v0

```

5. To visualize a trained `mjrl` agent's policy

```

MJPL python mabel/mabel_agents/mjrl/examine_policy.py -i mabel -e MableReachRandom-v0 -p 

```