[](https://pypi.org/project/torchx/)
[](LICENSE)
[](https://codecov.io/gh/pytorch/torchx)
# TorchX
TorchX is a universal job launcher for PyTorch applications.
TorchX is designed to have fast iteration time for training/research and support
for E2E production ML pipelines when you're ready.
TorchX currently supports:
* Kubernetes (EKS, GKE, AKS, etc)
* Slurm
* AWS Batch
* Docker
* Local
* Ray (prototype)
* GCP Batch (prototype)
Need a scheduler not listed? [Let us know!](https://github.com/pytorch/torchx/issues?q=is%3Aopen+is%3Aissue+label%3Ascheduler-request)
## Quickstart
See the [quickstart guide](https://pytorch.org/torchx/latest/quickstart.html).
## Documentation
* [Stable Documentation](https://pytorch.org/torchx/latest/)
* [Nightly Documentation](https://pytorch.org/torchx/main/)
## Requirements
torchx:
* python3 (3.8+)
* [PyTorch](https://pytorch.org/get-started/locally/)
* optional: [Docker](https://docs.docker.com/get-docker/) (needed for docker based schedulers)
Certain schedulers may require scheduler specific requirements. See installation
for info.
## Installation
### Stable
```bash
# install torchx sdk and CLI -- minimum dependencies
pip install torchx
# install torchx sdk and CLI -- all dependencies
pip install "torchx[dev]"
# install torchx kubeflow pipelines (kfp) support
pip install "torchx[kfp]"
# install torchx Kubernetes / Volcano support
pip install "torchx[kubernetes]"
# install torchx Ray support
pip install "torchx[ray]"
# install torchx GCP Batch support
pip install "torchx[gcp_batch]"
```
### Nightly
```bash
# install torchx sdk and CLI
pip install torchx-nightly[dev]
```
### Source
```bash
# install torchx sdk and CLI from source
$ pip install -e git+https://github.com/pytorch/torchx.git#egg=torchx
# install extra dependencies
$ pip install -e git+https://github.com/pytorch/torchx.git#egg=torchx[dev]
```
### Docker
TorchX provides a docker container for using as as part of a TorchX role.
See: https://github.com/pytorch/torchx/pkgs/container/torchx
## Contributing
We welcome PRs! See the [CONTRIBUTING](CONTRIBUTING.md) file.
## License
TorchX is BSD licensed, as found in the [LICENSE](LICENSE) file.
Raw data
{
"_id": null,
"home_page": "https://github.com/project-codeflare/torchx",
"name": "codeflare-torchx",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.7",
"maintainer_email": "",
"keywords": "pytorch,machine learning",
"author": "TorchX Devs",
"author_email": "torchx@fb.com",
"download_url": "",
"platform": null,
"description": "[](https://pypi.org/project/torchx/)\n[](LICENSE)\n\n\n[](https://codecov.io/gh/pytorch/torchx)\n\n\n# TorchX\n\n\nTorchX is a universal job launcher for PyTorch applications.\nTorchX is designed to have fast iteration time for training/research and support\nfor E2E production ML pipelines when you're ready.\n\nTorchX currently supports:\n\n* Kubernetes (EKS, GKE, AKS, etc)\n* Slurm\n* AWS Batch\n* Docker\n* Local\n* Ray (prototype)\n* GCP Batch (prototype)\n\nNeed a scheduler not listed? [Let us know!](https://github.com/pytorch/torchx/issues?q=is%3Aopen+is%3Aissue+label%3Ascheduler-request)\n\n## Quickstart\n\nSee the [quickstart guide](https://pytorch.org/torchx/latest/quickstart.html).\n\n## Documentation\n\n* [Stable Documentation](https://pytorch.org/torchx/latest/)\n* [Nightly Documentation](https://pytorch.org/torchx/main/)\n\n## Requirements\n\ntorchx:\n\n* python3 (3.8+)\n* [PyTorch](https://pytorch.org/get-started/locally/)\n* optional: [Docker](https://docs.docker.com/get-docker/) (needed for docker based schedulers)\n\nCertain schedulers may require scheduler specific requirements. See installation\nfor info.\n\n## Installation\n\n### Stable\n\n```bash\n# install torchx sdk and CLI -- minimum dependencies\npip install torchx\n\n# install torchx sdk and CLI -- all dependencies\npip install \"torchx[dev]\"\n\n# install torchx kubeflow pipelines (kfp) support\npip install \"torchx[kfp]\"\n\n# install torchx Kubernetes / Volcano support\npip install \"torchx[kubernetes]\"\n\n# install torchx Ray support\npip install \"torchx[ray]\"\n\n# install torchx GCP Batch support\npip install \"torchx[gcp_batch]\"\n```\n\n### Nightly\n\n```bash\n# install torchx sdk and CLI\npip install torchx-nightly[dev]\n```\n\n### Source\n\n```bash\n# install torchx sdk and CLI from source\n$ pip install -e git+https://github.com/pytorch/torchx.git#egg=torchx\n\n# install extra dependencies\n$ pip install -e git+https://github.com/pytorch/torchx.git#egg=torchx[dev]\n```\n\n### Docker\n\nTorchX provides a docker container for using as as part of a TorchX role.\n\nSee: https://github.com/pytorch/torchx/pkgs/container/torchx\n\n## Contributing\n\nWe welcome PRs! See the [CONTRIBUTING](CONTRIBUTING.md) file.\n\n## License\n\nTorchX is BSD licensed, as found in the [LICENSE](LICENSE) file.\n",
"bugtrack_url": null,
"license": "BSD-3",
"summary": "TorchX SDK and Components",
"version": "0.6.0.dev1",
"project_urls": {
"Homepage": "https://github.com/project-codeflare/torchx"
},
"split_keywords": [
"pytorch",
"machine learning"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "0c45bf809ebfa4ea7eb3ab3a3c4762179d2aca2834c459e29771e06d92a64244",
"md5": "49b1edb98b15da54daed019f6542d6af",
"sha256": "a03bf570c775cb7fc7787a436ef5a76c726669e287af510513c5bc03033d1fca"
},
"downloads": -1,
"filename": "codeflare_torchx-0.6.0.dev1-py3-none-any.whl",
"has_sig": false,
"md5_digest": "49b1edb98b15da54daed019f6542d6af",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.7",
"size": 244080,
"upload_time": "2023-09-08T22:22:12",
"upload_time_iso_8601": "2023-09-08T22:22:12.756198Z",
"url": "https://files.pythonhosted.org/packages/0c/45/bf809ebfa4ea7eb3ab3a3c4762179d2aca2834c459e29771e06d92a64244/codeflare_torchx-0.6.0.dev1-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-09-08 22:22:12",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "project-codeflare",
"github_project": "torchx",
"travis_ci": false,
"coveralls": true,
"github_actions": true,
"requirements": [],
"lcname": "codeflare-torchx"
}
JSON
