# Model Card Toolkit
The Model Card Toolkit (MCT) streamlines and automates generation of [Model Cards](https://modelcards.withgoogle.com/about) [1], machine learning documents that provide context and transparency into a model's development and performance. Integrating the MCT into your ML pipeline enables the sharing model metadata and metrics with researchers, developers, reporters, and more.
Some use cases of model cards include:
* Facilitating the exchange of information between model builders and product developers.
* Informing users of ML models to make better-informed decisions about how to use them (or how not to use them).
* Providing model information required for effective public oversight and accountability.
## Installation
The Model Card Toolkit is hosted on [PyPI](https://pypi.org/project/model-card-toolkit/), and can be installed with `pip install model-card-toolkit` (or `pip install model-card-toolkit
--use-deprecated=legacy-resolver` for versions of pip starting with 20.3). See [the installation guide](model_card_toolkit/documentation/guide/install.md) for more details.
## Getting Started
import model_card_toolkit
# Initialize the Model Card Toolkit with a path to store generate assets
model_card_output_path = ...
mct = model_card_toolkit.ModelCardToolkit(model_card_output_path)
# Initialize the model_card_toolkit.ModelCard, which can be freely populated
model_card = mct.scaffold_assets()
model_card.model_details.name = 'My Model'
# Write the model card data to a proto file
mct.update_model_card(model_card)
# Return the model card document as an HTML page
html = mct.export_format()
## Model Card Generation on TFX
If you are using [TensorFlow Extended (TFX)](https://www.tensorflow.org/tfx), you can
incorporate model card generation into your TFX pipeline via the `ModelCardGenerator`
component.
The `ModelCardGenerator` component is moving to the
[TFX Addons](https://github.com/tensorflow/tfx-addons) library and will no longer
be packaged in Model Card Toolkit from version 2.0.0. Before you can use the
component, you will need to install the `tfx-addons` package:
```sh
pip install tfx-addons[model_card_generator]
```
This page will be updated to include the new links for the Model Cards in TFX
guide and the end-to-end demo when the migration is completed.
## Schema
Model cards are stored in proto as an intermediate format. You can see the model card JSON schema in the `schema` directory.
## References
[1] https://arxiv.org/abs/1810.03993
Raw data
{
"_id": null,
"home_page": "https://github.com/tensorflow/model-card-toolkit",
"name": "model-card-toolkit",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.7,<4",
"maintainer_email": "",
"keywords": "model card toolkit ml metadata machine learning",
"author": "Google LLC",
"author_email": "tensorflow-extended-dev@googlegroups.com",
"download_url": "",
"platform": null,
"description": "# Model Card Toolkit\n\nThe Model Card Toolkit (MCT) streamlines and automates generation of [Model Cards](https://modelcards.withgoogle.com/about) [1], machine learning documents that provide context and transparency into a model's development and performance. Integrating the MCT into your ML pipeline enables the sharing model metadata and metrics with researchers, developers, reporters, and more.\n\nSome use cases of model cards include:\n\n* Facilitating the exchange of information between model builders and product developers.\n* Informing users of ML models to make better-informed decisions about how to use them (or how not to use them).\n* Providing model information required for effective public oversight and accountability.\n\n\n\n## Installation\n\nThe Model Card Toolkit is hosted on [PyPI](https://pypi.org/project/model-card-toolkit/), and can be installed with `pip install model-card-toolkit` (or `pip install model-card-toolkit\n--use-deprecated=legacy-resolver` for versions of pip starting with 20.3). See [the installation guide](model_card_toolkit/documentation/guide/install.md) for more details.\n\n## Getting Started\n\n import model_card_toolkit\n\n # Initialize the Model Card Toolkit with a path to store generate assets\n model_card_output_path = ...\n mct = model_card_toolkit.ModelCardToolkit(model_card_output_path)\n\n # Initialize the model_card_toolkit.ModelCard, which can be freely populated\n model_card = mct.scaffold_assets()\n model_card.model_details.name = 'My Model'\n\n # Write the model card data to a proto file\n mct.update_model_card(model_card)\n\n # Return the model card document as an HTML page\n html = mct.export_format()\n\n## Model Card Generation on TFX\n\nIf you are using [TensorFlow Extended (TFX)](https://www.tensorflow.org/tfx), you can\nincorporate model card generation into your TFX pipeline via the `ModelCardGenerator`\ncomponent.\n\nThe `ModelCardGenerator` component is moving to the\n[TFX Addons](https://github.com/tensorflow/tfx-addons) library and will no longer\nbe packaged in Model Card Toolkit from version 2.0.0. Before you can use the\ncomponent, you will need to install the `tfx-addons` package:\n\n```sh\npip install tfx-addons[model_card_generator]\n```\n\nThis page will be updated to include the new links for the Model Cards in TFX\nguide and the end-to-end demo when the migration is completed.\n\n## Schema\n\nModel cards are stored in proto as an intermediate format. You can see the model card JSON schema in the `schema` directory.\n\n## References\n\n[1] https://arxiv.org/abs/1810.03993\n\n\n",
"bugtrack_url": null,
"license": "Apache 2.0",
"summary": "Model Card Toolkit",
"version": "2.0.0",
"split_keywords": [
"model",
"card",
"toolkit",
"ml",
"metadata",
"machine",
"learning"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "e686e4c3f63cd6dbba95e22d56f4f5e486f3dc78762b3b053f7ee0b50bafd0c5",
"md5": "457907e5797ffad644a443200fce6651",
"sha256": "b195c4678529c6f14096b60ee263c81ff7c81c27773f90e81e55d06923d95844"
},
"downloads": -1,
"filename": "model_card_toolkit-2.0.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "457907e5797ffad644a443200fce6651",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.7,<4",
"size": 68829,
"upload_time": "2023-04-03T18:03:39",
"upload_time_iso_8601": "2023-04-03T18:03:39.087300Z",
"url": "https://files.pythonhosted.org/packages/e6/86/e4c3f63cd6dbba95e22d56f4f5e486f3dc78762b3b053f7ee0b50bafd0c5/model_card_toolkit-2.0.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-04-03 18:03:39",
"github": true,
"gitlab": false,
"bitbucket": false,
"github_user": "tensorflow",
"github_project": "model-card-toolkit",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "model-card-toolkit"
}
JSON
