# Zeroshot (Python)
Image classification for the masses
## Installation
Install via pip: `pip install zeroshot`
For GPU support, `pip install zeroshot[torch]`
N.B. In theory ONNX supports GPU, but the restrictions on CUDA version are iffy at best, and so for easiest results just use PyTorch. If you're brave, instead `pip install onnxruntime-gpu`.
## Usage
First, go to app.usezeroshot.com and create a classifier. Check out the video on the [landing page](usezeroshot.com) for an example.
Then, in Python (`image` should be an RGB numpy array with channels last):
```python
import zeroshot
# Create the classifier and preprocessing function.
classifier = zeroshot.Classifier("model-uuid-goes-here")
preprocess_fn = zeroshot.create_preprocess_fn()
# Run the model!
prediction = classifier.predict(preprocess_fn(image))
print(f"The image is class {prediction}")
```
You can also download the classifier and save it somewhere locally so you don't need to hit the server each time. Hit "download model" in the web-app and save the json file somewhere. You can then instead do:
```python
classifier = zeroshot.Classifier("/home/user/path/to/model.json")
```
## Additional Tips
* To use a GPU, install the torch backend with `pip install zeroshot[torch]`
* If you are hitting issues with torch trying to run on CPU, try disabling XFormers by setting XFORMERS_DISABLED=1 in your ENV varaibles.
## Read the docs
See the [docs](https://github.com/moonshinelabs-ai/zeroshot-docs/blob/main/general/getting_started.md) folder for some details on how things work under the hood.
## Get help
If you need help or just want to chat, join the [Moonshine Labs Slack server](https://join.slack.com/t/moonshinecommunity/shared_invite/zt-1rg1vnvmt-pleUR7TducaDiAhcmnqAQQ) and come hang out in the #zeroshot channel.
Raw data
{
"_id": null,
"home_page": "https://github.com/moonshinelabs/zeroshot-python",
"name": "zeroshot",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.10,<3.13",
"maintainer_email": "",
"keywords": "zeroshot,classifier,cv",
"author": "Nate Harada",
"author_email": "hello@usezeroshot.com",
"download_url": "https://files.pythonhosted.org/packages/d3/53/4ca17cfe405bce9e9d33b657c7d886d3c05cd942265e145af61026bb10f6/zeroshot-0.1.11.tar.gz",
"platform": null,
"description": "# Zeroshot (Python)\n\nImage classification for the masses\n\n## Installation\n\nInstall via pip: `pip install zeroshot`\n\nFor GPU support, `pip install zeroshot[torch]`\n\nN.B. In theory ONNX supports GPU, but the restrictions on CUDA version are iffy at best, and so for easiest results just use PyTorch. If you're brave, instead `pip install onnxruntime-gpu`.\n\n## Usage\n\nFirst, go to app.usezeroshot.com and create a classifier. Check out the video on the [landing page](usezeroshot.com) for an example.\n\nThen, in Python (`image` should be an RGB numpy array with channels last):\n\n```python\nimport zeroshot\n\n# Create the classifier and preprocessing function.\nclassifier = zeroshot.Classifier(\"model-uuid-goes-here\")\npreprocess_fn = zeroshot.create_preprocess_fn()\n\n# Run the model!\nprediction = classifier.predict(preprocess_fn(image))\nprint(f\"The image is class {prediction}\")\n```\n\nYou can also download the classifier and save it somewhere locally so you don't need to hit the server each time. Hit \"download model\" in the web-app and save the json file somewhere. You can then instead do:\n\n```python\nclassifier = zeroshot.Classifier(\"/home/user/path/to/model.json\")\n```\n\n## Additional Tips\n\n* To use a GPU, install the torch backend with `pip install zeroshot[torch]`\n* If you are hitting issues with torch trying to run on CPU, try disabling XFormers by setting XFORMERS_DISABLED=1 in your ENV varaibles.\n\n## Read the docs\n\nSee the [docs](https://github.com/moonshinelabs-ai/zeroshot-docs/blob/main/general/getting_started.md) folder for some details on how things work under the hood.\n\n## Get help\n\nIf you need help or just want to chat, join the [Moonshine Labs Slack server](https://join.slack.com/t/moonshinecommunity/shared_invite/zt-1rg1vnvmt-pleUR7TducaDiAhcmnqAQQ) and come hang out in the #zeroshot channel.\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "Image classifier with zero-shot learning.",
"version": "0.1.11",
"project_urls": {
"Homepage": "https://github.com/moonshinelabs/zeroshot-python",
"Repository": "https://github.com/moonshinelabs/zeroshot-python"
},
"split_keywords": [
"zeroshot",
"classifier",
"cv"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "01b8118650fa88ce84de4897229bce0026600657d497807fe4b066ea4727c1ef",
"md5": "3a1a8b8263405406f188d65f916e9e90",
"sha256": "208452598d4426b64f78f1c9c3d4caf0b314bc10a55657c38384c3f77d946b9f"
},
"downloads": -1,
"filename": "zeroshot-0.1.11-py3-none-any.whl",
"has_sig": false,
"md5_digest": "3a1a8b8263405406f188d65f916e9e90",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.10,<3.13",
"size": 26368,
"upload_time": "2023-12-05T00:52:13",
"upload_time_iso_8601": "2023-12-05T00:52:13.152005Z",
"url": "https://files.pythonhosted.org/packages/01/b8/118650fa88ce84de4897229bce0026600657d497807fe4b066ea4727c1ef/zeroshot-0.1.11-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "d3534ca17cfe405bce9e9d33b657c7d886d3c05cd942265e145af61026bb10f6",
"md5": "5283ea518438216de7f9c8529f62d323",
"sha256": "7e012ac58ccb1979560f294846d86c475d05f9bfe660e948e15294296b37522b"
},
"downloads": -1,
"filename": "zeroshot-0.1.11.tar.gz",
"has_sig": false,
"md5_digest": "5283ea518438216de7f9c8529f62d323",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.10,<3.13",
"size": 24388,
"upload_time": "2023-12-05T00:52:14",
"upload_time_iso_8601": "2023-12-05T00:52:14.962956Z",
"url": "https://files.pythonhosted.org/packages/d3/53/4ca17cfe405bce9e9d33b657c7d886d3c05cd942265e145af61026bb10f6/zeroshot-0.1.11.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-12-05 00:52:14",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "moonshinelabs",
"github_project": "zeroshot-python",
"github_not_found": true,
"lcname": "zeroshot"
}
JSON
