| Name | tyro JSON |
| Version |
0.9.27
JSON |
| download |
| home_page | None |
| Summary | CLI interfaces & config objects, from types |
| upload_time | 2025-07-29 22:29:50 |
| maintainer | None |
| docs_url | None |
| author | None |
| requires_python | >=3.8 |
| license | MIT |
| keywords |
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
<br />
<p align="center">
<!--
This README will be used for both GitHub and PyPI. We therefore:
- Keep all image URLs absolute.
- In the GitHub action we use for publishing, strip some HTML tags that aren't supported by PyPI.
-->
<img alt="tyro logo" src="https://brentyi.github.io/tyro/_static/logo-light.svg" width="200px" />
</p>
<p align="center">
<em><a href="https://brentyi.github.io/tyro">Documentation</a></em>
•
<em><code>pip install tyro</code></em>
</p>
<p align="center">
<!-- <img alt="build" src="https://github.com/brentyi/tyro/actions/workflows/build.yml/badge.svg" /> -->
<img alt="mypy" src="https://github.com/brentyi/tyro/actions/workflows/mypy.yml/badge.svg" />
<img alt="pyright" src="https://github.com/brentyi/tyro/actions/workflows/pyright.yml/badge.svg" />
<!-- <img alt="ruff" src="https://github.com/brentyi/tyro/actions/workflows/ruff.yml/badge.svg" /> -->
<a href="https://codecov.io/gh/brentyi/tyro">
<img alt="codecov" src="https://codecov.io/gh/brentyi/tyro/branch/main/graph/badge.svg" />
</a>
<a href="https://pypi.org/project/tyro/">
<img alt="codecov" src="https://img.shields.io/pypi/pyversions/tyro" />
</a>
</p>
<br />
<strong><code>tyro.cli()</code></strong> is a tool for generating CLI
interfaces from type-annotated Python.
We can define configurable scripts using functions:
```python
"""A command-line interface defined using a function signature.
Usage: python script_name.py --foo INT [--bar STR]
"""
import tyro
def main(foo: int, bar: str = "default") -> None:
... # Main body of a script.
if __name__ == "__main__":
# Generate a CLI and call `main` with its two arguments: `foo` and `bar`.
tyro.cli(main)
```
Or instantiate config objects defined using tools like `dataclasses`, `pydantic`, and `attrs`:
```python
"""A command-line interface defined using a class signature.
Usage: python script_name.py --foo INT [--bar STR]
"""
from dataclasses import dataclass
import tyro
@dataclass
class Config:
foo: int
bar: str = "default"
if __name__ == "__main__":
# Generate a CLI and instantiate `Config` with its two arguments: `foo` and `bar`.
config = tyro.cli(Config)
# Rest of script.
assert isinstance(config, Config) # Should pass.
```
Other features include helptext generation, nested structures, subcommands, and
shell completion. For examples and the API reference, see our
[documentation](https://brentyi.github.io/tyro).
### Why `tyro`?
1. **Define things once.** Standard Python type annotations, docstrings, and default values are parsed to automatically generate command-line interfaces with informative helptext.
2. **Static types.** Unlike tools dependent on dictionaries, YAML, or dynamic
namespaces, arguments populated by `tyro` benefit from IDE and language
server-supported operations — tab completion, rename, jump-to-def,
docstrings on hover — as well as static checking tools like `pyright` and
`mypy`.
3. **Modularity.** `tyro` supports hierarchical configuration structures, which
make it easy to decentralize definitions, defaults, and documentation.
### In the wild
`tyro` is designed to be lightweight enough for throwaway scripts, while
facilitating type safety and modularity for larger projects. Examples:
<table>
<tr>
<td>
<a href="https://github.com/nerfstudio-project/nerfstudio/">
nerfstudio-project/nerfstudio
<br /><img
alt="GitHub stars"
src="https://img.shields.io/github/stars/nerfstudio-project/nerfstudio?style=social"
/>
</a>
</td>
<td>
Open-source tools for neural radiance fields.
</td>
</tr>
<tr>
<td>
<a href="https://github.com/Sea-Snell/JAXSeq/">
Sea-Snell/JAXSeq
<br /><img
alt="GitHub stars"
src="https://img.shields.io/github/stars/Sea-Snell/JAXSeq?style=social"
/>
</a>
</td>
<td>Train very large language models in Jax.</td>
</tr>
<tr>
<td>
<a href="https://github.com/kevinzakka/obj2mjcf">
kevinzakka/obj2mjcf
<br /><img
alt="GitHub stars"
src="https://img.shields.io/github/stars/kevinzakka/obj2mjcf?style=social"
/>
</a>
</td>
<td>Interface for processing OBJ files for Mujoco.</td>
</tr>
<tr>
<td>
<a href="https://github.com/blurgyy/jaxngp">
blurgyy/jaxngp
<br /><img
alt="GitHub stars"
src="https://img.shields.io/github/stars/blurgyy/jaxngp?style=social"
/>
</a>
</td>
<td>
CUDA-accelerated implementation of
<a href="https://nvlabs.github.io/instant-ngp/">instant-ngp</a>, in JAX.
</td>
</tr>
<tr>
<td>
<a href="https://github.com/NVIDIAGameWorks/kaolin-wisp">
NVIDIAGameWorks/kaolin-wisp
<br /><img
alt="GitHub stars"
src="https://img.shields.io/github/stars/NVIDIAGameWorks/kaolin-wisp?style=social"
/>
</a>
</td>
<td>PyTorch library for neural fields.</td>
</tr>
<tr>
<td>
<a href="https://github.com/autonomousvision/sdfstudio">
autonomousvision/sdfstudio
<br /><img
alt="GitHub stars"
src="https://img.shields.io/github/stars/autonomousvision/sdfstudio?style=social"
/>
</a>
</td>
<td>Unified framework for surface reconstruction.</td>
</tr>
<tr>
<td>
<a href="https://github.com/openrlbenchmark/openrlbenchmark">
openrlbenchmark/openrlbenchmark
<br /><img
alt="GitHub stars"
src="https://img.shields.io/github/stars/openrlbenchmark/openrlbenchmark?style=social"
/>
</a>
</td>
<td>Collection of tracked experiments for reinforcement learning.</td>
</tr>
<tr>
<td>
<a href="https://github.com/vwxyzjn/cleanrl">
vwxyzjn/cleanrl
<br /><img
alt="GitHub stars"
src="https://img.shields.io/github/stars/vwxyzjn/cleanrl?style=social"
/>
</a>
</td>
<td>Single-file implementation of deep RL algorithms.</td>
</tr>
<tr>
<td>
<a href="https://github.com/pytorch-labs/LeanRL/">
pytorch-labs/LeanRL
<br /><img
alt="GitHub stars"
src="https://img.shields.io/github/stars/pytorch-labs/LeanRL?style=social"
/>
</a>
</td>
<td>Fork of CleanRL, optimized using PyTorch 2 features.</td>
</tr>
<tr>
<td>
<a href="https://github.com/pytorch/torchtitan">
pytorch/torchtitan
<br /><img
alt="GitHub stars"
src="https://img.shields.io/github/stars/pytorch/torchtitan?style=social"
/>
</a>
</td>
<td>PyTorch-native platform for training generative AI models.</td>
</tr>
<tr>
<td>
<a href="https://github.com/KwaiVGI/LivePortrait">
KwaiVGI/LivePortrait
<br /><img
alt="GitHub stars"
src="https://img.shields.io/github/stars/KwaiVGI/LivePortrait?style=social"
/>
</a>
</td>
<td>Stitching and retargeting for portraits.</td>
</tr>
<tr>
<td>
<a href="https://github.com/Physical-Intelligence/openpi/">
Physical-Intelligence/openpi
<br /><img
alt="GitHub stars"
src="https://img.shields.io/github/stars/Physical-Intelligence/openpi?style=social"
/>
</a>
</td>
<td>Open-source models for robotics.</td>
</tr>
<tr>
<td>
<a href="https://github.com/MalcolmMielle/bark_monitor">
MalcolmMielle/bark_monitor
<br /><img
alt="GitHub stars"
src="https://img.shields.io/github/stars/MalcolmMielle/bark_monitor?style=social"
/>
</a>
</td>
<td>Show your neighbor that your dog doesn't bark!</td>
</tr>
</table>
### Alternatives
`tyro` is an opinionated library. If any design decisions don't make sense,
feel free to file an issue!
You might also consider one of many alternative libraries. Some that we
particularly like:
- [cyclopts](https://github.com/BrianPugh/cyclopts) and
[defopt](https://defopt.readthedocs.io/), which have very comprehensive type
annotation support and a heavier emphasis on subcommand generation.
- [simple-parsing](https://github.com/lebrice/SimpleParsing) and
[jsonargparse](https://github.com/omni-us/jsonargparse), which provide deeper
integration with configuration file formats like YAML and JSON.
- [clipstick](https://github.com/sander76/clipstick), which focuses on
simplicity + generating CLIs from Pydantic models.
- [datargs](https://github.com/roee30/datargs), which provides a minimal API for
dataclasses.
- [fire](https://github.com/google/python-fire) and
[clize](https://github.com/epsy/clize), which support arguments without type
annotations.
We also have some notes on `tyro`'s design goals and other alternatives in the
docs [here](https://brentyi.github.io/tyro/goals_and_alternatives/).
Raw data
{
"_id": null,
"home_page": null,
"name": "tyro",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": null,
"keywords": null,
"author": null,
"author_email": "brentyi <brentyi@berkeley.edu>",
"download_url": "https://files.pythonhosted.org/packages/48/4b/c2b5e9b497bdd03fbf78f1fb83da621e6609d6a764ea0c34f9486dcc3e95/tyro-0.9.27.tar.gz",
"platform": null,
"description": "<br />\n<p align=\"center\">\n <!--\n This README will be used for both GitHub and PyPI. We therefore:\n - Keep all image URLs absolute.\n - In the GitHub action we use for publishing, strip some HTML tags that aren't supported by PyPI.\n -->\n <img alt=\"tyro logo\" src=\"https://brentyi.github.io/tyro/_static/logo-light.svg\" width=\"200px\" />\n\n</p>\n\n<p align=\"center\">\n <em><a href=\"https://brentyi.github.io/tyro\">Documentation</a></em>\n • \n <em><code>pip install tyro</code></em>\n</p>\n\n<p align=\"center\">\n <!-- <img alt=\"build\" src=\"https://github.com/brentyi/tyro/actions/workflows/build.yml/badge.svg\" /> -->\n <img alt=\"mypy\" src=\"https://github.com/brentyi/tyro/actions/workflows/mypy.yml/badge.svg\" />\n <img alt=\"pyright\" src=\"https://github.com/brentyi/tyro/actions/workflows/pyright.yml/badge.svg\" />\n <!-- <img alt=\"ruff\" src=\"https://github.com/brentyi/tyro/actions/workflows/ruff.yml/badge.svg\" /> -->\n <a href=\"https://codecov.io/gh/brentyi/tyro\">\n <img alt=\"codecov\" src=\"https://codecov.io/gh/brentyi/tyro/branch/main/graph/badge.svg\" />\n </a>\n <a href=\"https://pypi.org/project/tyro/\">\n <img alt=\"codecov\" src=\"https://img.shields.io/pypi/pyversions/tyro\" />\n </a>\n</p>\n\n<br />\n\n<strong><code>tyro.cli()</code></strong> is a tool for generating CLI\ninterfaces from type-annotated Python.\n\nWe can define configurable scripts using functions:\n\n```python\n\"\"\"A command-line interface defined using a function signature.\n\nUsage: python script_name.py --foo INT [--bar STR]\n\"\"\"\n\nimport tyro\n\ndef main(foo: int, bar: str = \"default\") -> None:\n ... # Main body of a script.\n\nif __name__ == \"__main__\":\n # Generate a CLI and call `main` with its two arguments: `foo` and `bar`.\n tyro.cli(main)\n```\n\nOr instantiate config objects defined using tools like `dataclasses`, `pydantic`, and `attrs`:\n\n```python\n\"\"\"A command-line interface defined using a class signature.\n\nUsage: python script_name.py --foo INT [--bar STR]\n\"\"\"\n\nfrom dataclasses import dataclass\nimport tyro\n\n@dataclass\nclass Config:\n foo: int\n bar: str = \"default\"\n\nif __name__ == \"__main__\":\n # Generate a CLI and instantiate `Config` with its two arguments: `foo` and `bar`.\n config = tyro.cli(Config)\n\n # Rest of script.\n assert isinstance(config, Config) # Should pass.\n```\n\nOther features include helptext generation, nested structures, subcommands, and\nshell completion. For examples and the API reference, see our\n[documentation](https://brentyi.github.io/tyro).\n\n### Why `tyro`?\n\n1. **Define things once.** Standard Python type annotations, docstrings, and default values are parsed to automatically generate command-line interfaces with informative helptext.\n\n2. **Static types.** Unlike tools dependent on dictionaries, YAML, or dynamic\n namespaces, arguments populated by `tyro` benefit from IDE and language\n server-supported operations \u2014 tab completion, rename, jump-to-def,\n docstrings on hover \u2014 as well as static checking tools like `pyright` and\n `mypy`.\n\n3. **Modularity.** `tyro` supports hierarchical configuration structures, which\n make it easy to decentralize definitions, defaults, and documentation.\n\n### In the wild\n\n`tyro` is designed to be lightweight enough for throwaway scripts, while\nfacilitating type safety and modularity for larger projects. Examples:\n\n<table>\n <tr>\n <td>\n <a href=\"https://github.com/nerfstudio-project/nerfstudio/\">\n nerfstudio-project/nerfstudio\n <br /><img\n alt=\"GitHub stars\"\n src=\"https://img.shields.io/github/stars/nerfstudio-project/nerfstudio?style=social\"\n />\n </a>\n </td>\n <td>\n Open-source tools for neural radiance fields.\n </td>\n </tr>\n <tr>\n <td>\n <a href=\"https://github.com/Sea-Snell/JAXSeq/\">\n Sea-Snell/JAXSeq\n <br /><img\n alt=\"GitHub stars\"\n src=\"https://img.shields.io/github/stars/Sea-Snell/JAXSeq?style=social\"\n />\n </a>\n </td>\n <td>Train very large language models in Jax.</td>\n </tr>\n <tr>\n <td>\n <a href=\"https://github.com/kevinzakka/obj2mjcf\">\n kevinzakka/obj2mjcf\n <br /><img\n alt=\"GitHub stars\"\n src=\"https://img.shields.io/github/stars/kevinzakka/obj2mjcf?style=social\"\n />\n </a>\n </td>\n <td>Interface for processing OBJ files for Mujoco.</td>\n </tr>\n <tr>\n <td>\n <a href=\"https://github.com/blurgyy/jaxngp\">\n blurgyy/jaxngp\n <br /><img\n alt=\"GitHub stars\"\n src=\"https://img.shields.io/github/stars/blurgyy/jaxngp?style=social\"\n />\n </a>\n </td>\n <td>\n CUDA-accelerated implementation of\n <a href=\"https://nvlabs.github.io/instant-ngp/\">instant-ngp</a>, in JAX.\n </td>\n </tr>\n <tr>\n <td>\n <a href=\"https://github.com/NVIDIAGameWorks/kaolin-wisp\">\n NVIDIAGameWorks/kaolin-wisp\n <br /><img\n alt=\"GitHub stars\"\n src=\"https://img.shields.io/github/stars/NVIDIAGameWorks/kaolin-wisp?style=social\"\n />\n </a>\n </td>\n <td>PyTorch library for neural fields.</td>\n </tr>\n <tr>\n <td>\n <a href=\"https://github.com/autonomousvision/sdfstudio\">\n autonomousvision/sdfstudio\n <br /><img\n alt=\"GitHub stars\"\n src=\"https://img.shields.io/github/stars/autonomousvision/sdfstudio?style=social\"\n />\n </a>\n </td>\n <td>Unified framework for surface reconstruction.</td>\n </tr>\n <tr>\n <td>\n <a href=\"https://github.com/openrlbenchmark/openrlbenchmark\">\n openrlbenchmark/openrlbenchmark\n <br /><img\n alt=\"GitHub stars\"\n src=\"https://img.shields.io/github/stars/openrlbenchmark/openrlbenchmark?style=social\"\n />\n </a>\n </td>\n <td>Collection of tracked experiments for reinforcement learning.</td>\n </tr>\n <tr>\n <td>\n <a href=\"https://github.com/vwxyzjn/cleanrl\">\n vwxyzjn/cleanrl\n <br /><img\n alt=\"GitHub stars\"\n src=\"https://img.shields.io/github/stars/vwxyzjn/cleanrl?style=social\"\n />\n </a>\n </td>\n <td>Single-file implementation of deep RL algorithms.</td>\n </tr>\n <tr>\n <td>\n <a href=\"https://github.com/pytorch-labs/LeanRL/\">\n pytorch-labs/LeanRL\n <br /><img\n alt=\"GitHub stars\"\n src=\"https://img.shields.io/github/stars/pytorch-labs/LeanRL?style=social\"\n />\n </a>\n </td>\n <td>Fork of CleanRL, optimized using PyTorch 2 features.</td>\n </tr>\n <tr>\n <td>\n <a href=\"https://github.com/pytorch/torchtitan\">\n pytorch/torchtitan\n <br /><img\n alt=\"GitHub stars\"\n src=\"https://img.shields.io/github/stars/pytorch/torchtitan?style=social\"\n />\n </a>\n </td>\n <td>PyTorch-native platform for training generative AI models.</td>\n </tr>\n <tr>\n <td>\n <a href=\"https://github.com/KwaiVGI/LivePortrait\">\n KwaiVGI/LivePortrait\n <br /><img\n alt=\"GitHub stars\"\n src=\"https://img.shields.io/github/stars/KwaiVGI/LivePortrait?style=social\"\n />\n </a>\n </td>\n <td>Stitching and retargeting for portraits.</td>\n </tr>\n <tr>\n <td>\n <a href=\"https://github.com/Physical-Intelligence/openpi/\">\n Physical-Intelligence/openpi\n <br /><img\n alt=\"GitHub stars\"\n src=\"https://img.shields.io/github/stars/Physical-Intelligence/openpi?style=social\"\n />\n </a>\n </td>\n <td>Open-source models for robotics.</td>\n </tr>\n <tr>\n <td>\n <a href=\"https://github.com/MalcolmMielle/bark_monitor\">\n MalcolmMielle/bark_monitor\n <br /><img\n alt=\"GitHub stars\"\n src=\"https://img.shields.io/github/stars/MalcolmMielle/bark_monitor?style=social\"\n />\n </a>\n </td>\n <td>Show your neighbor that your dog doesn't bark!</td>\n </tr>\n</table>\n\n### Alternatives\n\n`tyro` is an opinionated library. If any design decisions don't make sense,\nfeel free to file an issue!\n\nYou might also consider one of many alternative libraries. Some that we\nparticularly like:\n\n- [cyclopts](https://github.com/BrianPugh/cyclopts) and\n [defopt](https://defopt.readthedocs.io/), which have very comprehensive type\n annotation support and a heavier emphasis on subcommand generation.\n- [simple-parsing](https://github.com/lebrice/SimpleParsing) and\n [jsonargparse](https://github.com/omni-us/jsonargparse), which provide deeper\n integration with configuration file formats like YAML and JSON.\n- [clipstick](https://github.com/sander76/clipstick), which focuses on\n simplicity + generating CLIs from Pydantic models.\n- [datargs](https://github.com/roee30/datargs), which provides a minimal API for\n dataclasses.\n- [fire](https://github.com/google/python-fire) and\n [clize](https://github.com/epsy/clize), which support arguments without type\n annotations.\n\nWe also have some notes on `tyro`'s design goals and other alternatives in the\ndocs [here](https://brentyi.github.io/tyro/goals_and_alternatives/).\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "CLI interfaces & config objects, from types",
"version": "0.9.27",
"project_urls": {
"GitHub": "https://github.com/brentyi/tyro"
},
"split_keywords": [],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "36ef98b2700c6a262a9d78eaec5b16916a75a63f7c1e642cfce0717c440d2f9b",
"md5": "96bfe641626f01491c702813a7bb572c",
"sha256": "f51655c45be6ba297af47cfc04622287422177448a060ffbec0f5fa905046f41"
},
"downloads": -1,
"filename": "tyro-0.9.27-py3-none-any.whl",
"has_sig": false,
"md5_digest": "96bfe641626f01491c702813a7bb572c",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 129003,
"upload_time": "2025-07-29T22:29:48",
"upload_time_iso_8601": "2025-07-29T22:29:48.629461Z",
"url": "https://files.pythonhosted.org/packages/36/ef/98b2700c6a262a9d78eaec5b16916a75a63f7c1e642cfce0717c440d2f9b/tyro-0.9.27-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "484bc2b5e9b497bdd03fbf78f1fb83da621e6609d6a764ea0c34f9486dcc3e95",
"md5": "8ebb66da16507ac6a8b6f8538986a0fb",
"sha256": "f7b16340bc07b1eeb0a06880c9fcdddf0cfd084fbad40baf3072361c5a63b268"
},
"downloads": -1,
"filename": "tyro-0.9.27.tar.gz",
"has_sig": false,
"md5_digest": "8ebb66da16507ac6a8b6f8538986a0fb",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 307477,
"upload_time": "2025-07-29T22:29:50",
"upload_time_iso_8601": "2025-07-29T22:29:50.018449Z",
"url": "https://files.pythonhosted.org/packages/48/4b/c2b5e9b497bdd03fbf78f1fb83da621e6609d6a764ea0c34f9486dcc3e95/tyro-0.9.27.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-07-29 22:29:50",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "brentyi",
"github_project": "tyro",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "tyro"
}
