onnx-donut

Name	onnx-donut JSON
Version	0.1.0 JSON
	download
home_page
Summary	Export Donut model to onnx and run it with onnxruntime
upload_time	2023-11-20 14:39:29
maintainer
docs_url	None
author
requires_python	>=3.9
license
keywords	donut onnx
VCS
bugtrack_url
requirements	No requirements were recorded.
Travis-CI	No Travis.
coveralls test coverage	No coveralls.

            # Onnx Donut

Package to export a [Donut](https://github.com/clovaai/donut) model from pytorch to ONNX, then run it with onnxruntime.

## Installation

```shell
pip install onnx-donut
```

## Export to onnx

```python
from onnx_donut.exporter import export_onnx
from onnx_donut.quantizer import quantize

# Hugging Face model card or folder
model_path = "naver-clova-ix/donut-base-finetuned-docvqa"

# Folder where the exported model will be stored
dst_folder = "converted_donut"

# Export from Pytorch to ONNX
export_onnx(model_path, dst_folder, opset_version=16)

# Quantize your model to int8
quantize(dst_folder, dst_folder + "_quant")

```

## Model inference with onnxruntime

```python
from onnx_donut.predictor import OnnxPredictor
import numpy as np
from PIL import Image

# Image path to run on
img_path = "/path/to/your/image.png"

# Folder where the exported model will be stored
onnx_folder = "converted_donut"

# Read image
img = np.array(Image.open(img_path).convert('RGB'))

# Instantiate ONNX predictor
predictor = OnnxPredictor(model_folder=onnx_folder, sess_options=options)

# Write your prompt accordingly to the model you use
prompt = f"<s_docvqa><s_question>what is the title?</s_question><s_answer>"

# Run prediction
out = predictor.generate(img, prompt)

# Display prediction
print(out)
```

            

Raw data

            {
    "_id": null,
    "home_page": "",
    "name": "onnx-donut",
    "maintainer": "",
    "docs_url": null,
    "requires_python": ">=3.9",
    "maintainer_email": "",
    "keywords": "donut,onnx",
    "author": "",
    "author_email": "Ambroise Berthe <ambroise.berthe@ikomia.ai>",
    "download_url": "",
    "platform": null,
    "description": "# Onnx Donut\n\nPackage to export a [Donut](https://github.com/clovaai/donut) model from pytorch to ONNX, then run it with onnxruntime.\n\n## Installation\n\n```shell\npip install onnx-donut\n```\n\n## Export to onnx\n\n```python\nfrom onnx_donut.exporter import export_onnx\nfrom onnx_donut.quantizer import quantize\n\n# Hugging Face model card or folder\nmodel_path = \"naver-clova-ix/donut-base-finetuned-docvqa\"\n\n# Folder where the exported model will be stored\ndst_folder = \"converted_donut\"\n\n# Export from Pytorch to ONNX\nexport_onnx(model_path, dst_folder, opset_version=16)\n\n# Quantize your model to int8\nquantize(dst_folder, dst_folder + \"_quant\")\n\n```\n\n## Model inference with onnxruntime\n\n```python\nfrom onnx_donut.predictor import OnnxPredictor\nimport numpy as np\nfrom PIL import Image\n\n# Image path to run on\nimg_path = \"/path/to/your/image.png\"\n\n# Folder where the exported model will be stored\nonnx_folder = \"converted_donut\"\n\n# Read image\nimg = np.array(Image.open(img_path).convert('RGB'))\n\n# Instantiate ONNX predictor\npredictor = OnnxPredictor(model_folder=onnx_folder, sess_options=options)\n\n# Write your prompt accordingly to the model you use\nprompt = f\"<s_docvqa><s_question>what is the title?</s_question><s_answer>\"\n\n# Run prediction\nout = predictor.generate(img, prompt)\n\n# Display prediction\nprint(out)\n```\n",
    "bugtrack_url": null,
    "license": "",
    "summary": "Export Donut model to onnx and run it with onnxruntime",
    "version": "0.1.0",
    "project_urls": null,
    "split_keywords": [
        "donut",
        "onnx"
    ],
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "blake2b_256": "61b4451832aa4ee719e68b6d155b7544f502a3c4b0859fe2b03020f9a1b87ca6",
                "md5": "776de58e1fd007355bacb68c1bd038a7",
                "sha256": "e4bc5811952ccd82b2efda75ec02a1fb7680a099ff41d0629c8c2a8ecbd7c6e7"
            },
            "downloads": -1,
            "filename": "onnx_donut-0.1.0-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "776de58e1fd007355bacb68c1bd038a7",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": ">=3.9",
            "size": 10714,
            "upload_time": "2023-11-20T14:39:29",
            "upload_time_iso_8601": "2023-11-20T14:39:29.888395Z",
            "url": "https://files.pythonhosted.org/packages/61/b4/451832aa4ee719e68b6d155b7544f502a3c4b0859fe2b03020f9a1b87ca6/onnx_donut-0.1.0-py3-none-any.whl",
            "yanked": false,
            "yanked_reason": null
        }
    ],
    "upload_time": "2023-11-20 14:39:29",
    "github": false,
    "gitlab": false,
    "bitbucket": false,
    "codeberg": false,
    "lcname": "onnx-donut"
}