<div align="center">
<div align="center">
<img src="docs/assets/pytorch-boost-banner.svg" width="300px" style="max-width: 100%;">
</div>
**Deep Learning framework to accelerate the training of PyTorch models.**
______________________________________________________________________
<p align="center">
<a href="https://www.pytorch-boost.ai/">Boost</a> •
<a href="https://pytorch-boost.readthedocs.io/en/stable/">Docs</a> •
<a href="#community">Community</a> •
<a href="https://kswain55.github.io/pytorch-boost/stable/generated/CONTRIBUTING.html">Contribute</a>
</p>
PyTorch-Boost is a deep learning optimization library, offering a lightweight and user-friendly wrapper for custom CUDA functions. Designed to accelerate PyTorch model training and inference, PyTorch-Boost facilitates the application of high-precision and low-precision modalities— encompassing 32-bit, 16-bit, 8-bit, 4-bit, and mixed-precision training—across multiple GPUs or machines with its distributed training capabilities.
For detailed usage, installation guidelines, and more, please refer to our comprehensive documentation.
</div>
______________________________________________________________________
## Install PyTorch-Boost
Simple installation from PyPI
```bash
pip install pytorch-boost
```
<details>
<summary>Other installation options</summary>
#### Install with optional dependencies
```bash
pip install pytorch-boost['extra']
```
#### Conda
```bash
conda install pytorch-boost -c conda-forge
```
#### Install stable version
```bash
pip install https://github.com/kswain55/pytorch-boost/archive/refs/heads/release/stable.zip -U
```
#### Install bleeding-edge
```bash
pip install https://github.com/kswain55/pytorch-boost/archive/refs/heads/master.zip -U
```
</details>
______________________________________________________________________
## Compatibility
Supported Platform:
* Linux
Supported Hardware:
* NVIDIA Hopper, Ada, Ampere, Volta, and Turing architecture
* AMD CDNA 2/3 and RDNA 2/3
Supported Software Layer:
* CUDA 11.8 - 12.2 (NVIDIA)
* ROCM 5.3 and HIP compilation tools (AMD)
______________________________________________________________________
## Acknowledgements
We would like to thank Tim Dettmers for his work [bitsandbytes](https://github.com/TimDettmers/bitsandbytes) and the team developing [PyTorch](https://github.com/pytorch/pytorch), allowing us to make this possible. We would also like to thank NVIDIA and AMD for providng the neccessary hardware for testing.
______________________________________________________________________
## How to cite us
If you found this library useful, please consider citing our work:
```bibtex
@article{boost2023pytorch,
title={Boost},
author={minyoung huh, kswain},
journal={arXiv preprint arXiv:2023.2023},
year={2023}
}
```
Raw data
{
"_id": null,
"home_page": "https://github.com/kswain55/pytorch-boost",
"name": "pytorch-boost",
"maintainer": "",
"docs_url": null,
"requires_python": "",
"maintainer_email": "",
"keywords": "artificial intelligence",
"author": "minyoung huh, kswain",
"author_email": "minhuh@mit.edu, kswain@mit.edu",
"download_url": "",
"platform": null,
"description": "<div align=\"center\">\n\n<div align=\"center\">\n <img src=\"docs/assets/pytorch-boost-banner.svg\" width=\"300px\" style=\"max-width: 100%;\">\n</div>\n\n**Deep Learning framework to accelerate the training of PyTorch models.**\n\n______________________________________________________________________\n\n<p align=\"center\">\n <a href=\"https://www.pytorch-boost.ai/\">Boost</a> \u2022\n <a href=\"https://pytorch-boost.readthedocs.io/en/stable/\">Docs</a> \u2022\n <a href=\"#community\">Community</a> \u2022\n <a href=\"https://kswain55.github.io/pytorch-boost/stable/generated/CONTRIBUTING.html\">Contribute</a>\n</p>\n\n\nPyTorch-Boost is a deep learning optimization library, offering a lightweight and user-friendly wrapper for custom CUDA functions. Designed to accelerate PyTorch model training and inference, PyTorch-Boost facilitates the application of high-precision and low-precision modalities\u2014 encompassing 32-bit, 16-bit, 8-bit, 4-bit, and mixed-precision training\u2014across multiple GPUs or machines with its distributed training capabilities.\n\nFor detailed usage, installation guidelines, and more, please refer to our comprehensive documentation.\n\n</div>\n\n______________________________________________________________________\n\n## Install PyTorch-Boost\n\nSimple installation from PyPI\n\n```bash\npip install pytorch-boost\n```\n\n<details>\n <summary>Other installation options</summary>\n\n#### Install with optional dependencies\n\n```bash\npip install pytorch-boost['extra']\n```\n\n#### Conda\n\n```bash\nconda install pytorch-boost -c conda-forge\n```\n\n#### Install stable version\n\n```bash\npip install https://github.com/kswain55/pytorch-boost/archive/refs/heads/release/stable.zip -U\n```\n\n#### Install bleeding-edge\n\n```bash\npip install https://github.com/kswain55/pytorch-boost/archive/refs/heads/master.zip -U\n```\n</details>\n\n______________________________________________________________________\n\n\n## Compatibility\nSupported Platform:\n* Linux\n\nSupported Hardware:\n* NVIDIA Hopper, Ada, Ampere, Volta, and Turing architecture\n* AMD CDNA 2/3 and RDNA 2/3\n\nSupported Software Layer:\n* CUDA 11.8 - 12.2 (NVIDIA)\n* ROCM 5.3 and HIP compilation tools (AMD) \n\n______________________________________________________________________\n\n## Acknowledgements\nWe would like to thank Tim Dettmers for his work [bitsandbytes](https://github.com/TimDettmers/bitsandbytes) and the team developing [PyTorch](https://github.com/pytorch/pytorch), allowing us to make this possible. We would also like to thank NVIDIA and AMD for providng the neccessary hardware for testing. \n\n______________________________________________________________________\n\n\n## How to cite us\nIf you found this library useful, please consider citing our work:\n\n```bibtex\n@article{boost2023pytorch,\n title={Boost},\n author={minyoung huh, kswain},\n journal={arXiv preprint arXiv:2023.2023},\n year={2023}\n}\n```\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Deep learning optimization library for PyTorch",
"version": "0.0.1",
"project_urls": {
"Homepage": "https://github.com/kswain55/pytorch-boost"
},
"split_keywords": [
"artificial",
"intelligence"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "c4ad7519f95eede07de6718a2f290670bf746b00355e442fe5af3c3414aacbdf",
"md5": "0b46eb25a185ab267da8cf01088accda",
"sha256": "e56ef54d28b19479d04dcdeb16ce0a38d84a2f3968de3e9356b096c98a89da9a"
},
"downloads": -1,
"filename": "pytorch_boost-0.0.1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "0b46eb25a185ab267da8cf01088accda",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 3303,
"upload_time": "2023-07-14T14:46:45",
"upload_time_iso_8601": "2023-07-14T14:46:45.764918Z",
"url": "https://files.pythonhosted.org/packages/c4/ad/7519f95eede07de6718a2f290670bf746b00355e442fe5af3c3414aacbdf/pytorch_boost-0.0.1-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-07-14 14:46:45",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "kswain55",
"github_project": "pytorch-boost",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "pytorch-boost"
}
JSON
