# BreedGym
## Installation
Using pip:
```bash
pip install breedgym
```
From source:
```bash
git clone https://github.com/younik/breedgym
cd breedgym
pip install -e .
```
## Quickstart
BreedGym environments implement the [Gymnasium](https://gymnasium.farama.org) API, making it easy to use it with your preferred learning library.
```python
import gymnasium as gym
import numpy as np
env = gym.make(
"breedgym:BreedGym",
genetic_map="path/to/genetic_map.txt",
initial_population="path/to/geno.npy",
num_generations=10
)
print("Observation space:", env.observation_space)
print("Action space:", env.action_space)
```
To test, you can use the sample data we provide [here](https://github.com/younik/breedgym/tree/main/breedgym/data). In the case of the small sample data, we have 370 initial population members with 10k markers.
```
Observation space: Box(False, True, (370, 10000, 2), bool)
Action space: Sequence(Tuple(Discrete(370), Discrete(370)), stack=False)
```
After initializing the environment, we can interact with it as a standard Gymnasium environment:
```python
initial_pop, info = env.reset()
tru = False
for gen_number in range(10):
assert not tru
act = env.action_space.sample()
pop, rew, ter, tru, infos = env.step(np.asarray(act))
```
After 10 generations, we expect the environment to truncate, as we specified 10 generations horizon during environment initialization:
```python
assert tru
print("Reward (GEBV mean):", rew)
```
The full list of environments can be found [here](https://github.com/younik/breedgym/blob/main/breedgym/__init__.py).
## Citing
```
@inproceedings{younis2023breeding,
title={Breeding Programs Optimization with Reinforcement Learning},
author={Younis, Omar G. and Corinzia, Luca and Athanasiadis, Ioannis N and Krause, Andreas and Buhmann, Joachim and Turchetta, Matteo},
booktitle={NeurIPS 2023 Workshop on Tackling Climate Change with Machine Learning},
url={https://www.climatechange.ai/papers/neurips2023/93},
year={2023}
}
```
Raw data
{
"_id": null,
"home_page": null,
"name": "breedgym",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.9",
"maintainer_email": null,
"keywords": "Reinforcement Learning, Breeding, genetics, bioinformatics",
"author": null,
"author_email": "\"Omar G. Younis\" <omar.g.younis@gmail.com>",
"download_url": "https://files.pythonhosted.org/packages/24/72/ef5d12cb412e3004fc36a43dbcc9149c427e8c6eed8bc6c9499baa41bccc/breedgym-0.0.2a0.tar.gz",
"platform": null,
"description": "# BreedGym\n\n## Installation\n\nUsing pip:\n```bash\npip install breedgym\n```\n\nFrom source:\n```bash\ngit clone https://github.com/younik/breedgym\ncd breedgym\npip install -e .\n```\n\n## Quickstart\n\nBreedGym environments implement the [Gymnasium](https://gymnasium.farama.org) API, making it easy to use it with your preferred learning library.\n\n```python\nimport gymnasium as gym\nimport numpy as np\n\nenv = gym.make(\n \"breedgym:BreedGym\",\n genetic_map=\"path/to/genetic_map.txt\",\n initial_population=\"path/to/geno.npy\",\n num_generations=10\n)\nprint(\"Observation space:\", env.observation_space)\nprint(\"Action space:\", env.action_space)\n```\nTo test, you can use the sample data we provide [here](https://github.com/younik/breedgym/tree/main/breedgym/data). In the case of the small sample data, we have 370 initial population members with 10k markers. \n\n```\nObservation space: Box(False, True, (370, 10000, 2), bool)\nAction space: Sequence(Tuple(Discrete(370), Discrete(370)), stack=False)\n```\n\nAfter initializing the environment, we can interact with it as a standard Gymnasium environment:\n\n```python\ninitial_pop, info = env.reset()\ntru = False\nfor gen_number in range(10):\n assert not tru\n act = env.action_space.sample()\n pop, rew, ter, tru, infos = env.step(np.asarray(act))\n```\n\nAfter 10 generations, we expect the environment to truncate, as we specified 10 generations horizon during environment initialization:\n\n```python\nassert tru\nprint(\"Reward (GEBV mean):\", rew)\n```\n\nThe full list of environments can be found [here](https://github.com/younik/breedgym/blob/main/breedgym/__init__.py).\n\n## Citing\n\n```\n@inproceedings{younis2023breeding,\n title={Breeding Programs Optimization with Reinforcement Learning},\n author={Younis, Omar G. and Corinzia, Luca and Athanasiadis, Ioannis N and Krause, Andreas and Buhmann, Joachim and Turchetta, Matteo},\n booktitle={NeurIPS 2023 Workshop on Tackling Climate Change with Machine Learning},\n url={https://www.climatechange.ai/papers/neurips2023/93},\n year={2023}\n}\n```\n",
"bugtrack_url": null,
"license": "BSD-3-Clause",
"summary": "Suite of Gymnasium environments for optimizing breeding programs",
"version": "0.0.2a0",
"project_urls": {
"Bug Report": "https://github.com/younik/breedgym/issues",
"Repository": "https://github.com/younik/breedgym"
},
"split_keywords": [
"reinforcement learning",
" breeding",
" genetics",
" bioinformatics"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "49358bd8c67219f860c1ceacc3d5238e9c382801778a474ffa1fa3c4efb19282",
"md5": "9b5806283a0276d33dc508ef17587875",
"sha256": "23faabaf12e0a376d3b98d538da11af1499afc79cef835d3951c3147f65f4895"
},
"downloads": -1,
"filename": "breedgym-0.0.2a0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "9b5806283a0276d33dc508ef17587875",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.9",
"size": 652057,
"upload_time": "2024-05-26T18:11:15",
"upload_time_iso_8601": "2024-05-26T18:11:15.123456Z",
"url": "https://files.pythonhosted.org/packages/49/35/8bd8c67219f860c1ceacc3d5238e9c382801778a474ffa1fa3c4efb19282/breedgym-0.0.2a0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "2472ef5d12cb412e3004fc36a43dbcc9149c427e8c6eed8bc6c9499baa41bccc",
"md5": "1f5ec916e760ee5b794b8294d6e48c6c",
"sha256": "db16261de673c1fb68556f49e4feef892ad3b894d2363c7f6450dd3de722dbd2"
},
"downloads": -1,
"filename": "breedgym-0.0.2a0.tar.gz",
"has_sig": false,
"md5_digest": "1f5ec916e760ee5b794b8294d6e48c6c",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.9",
"size": 562175,
"upload_time": "2024-05-26T18:11:18",
"upload_time_iso_8601": "2024-05-26T18:11:18.630716Z",
"url": "https://files.pythonhosted.org/packages/24/72/ef5d12cb412e3004fc36a43dbcc9149c427e8c6eed8bc6c9499baa41bccc/breedgym-0.0.2a0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-05-26 18:11:18",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "younik",
"github_project": "breedgym",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "breedgym"
}
JSON
