# Habana TensorFlow Python package
`habana-tensorflow` is a package designed for easy integration of Gaudi bridge to Tensorflow topologies.
Main entry point (Python import) is `habana_frameworks.tensorflow` module that contains all the necessary functions to work with Gaudi.
## Structure
Properly built wheel contains:
- `habana_frameworks` python namespace (with all the folder structure inside)
- libraries for all supported versions of TF (see `habana_frameworks.tensorflow.sysconfig._supported_tf_versions`) located in `habana_frameworks/tensorflow/tf<version_with_underscores>/lib/habanalabs/`
- proper licensing
When loading `habana_tensorflow`, there is sanity check for currently installed TF version in the system, to match libraries found in the module.
## TensorFlow package (habana_frameworks.tensorflow)
Most important module inside is ``library_loader`` and function `load_habana_module()`, that is exposed directly after import.
It initializes the module properly.
Example:
```python
import habana_frameworks.tensorflow as htf
htf.load_habana_module()
```
### List of modules inside ``habana_frameworks.tensorflow``
All modules are described in [official docs](https://docs.habana.ai/en/latest/Tensorflow_User_Guide/Tensorflow_User_Guide.html#python-package-habana-frameworks-tensorflow).
Raw data
{
"_id": null,
"home_page": "https://habana.ai/",
"name": "habana-tensorflow",
"maintainer": "",
"docs_url": null,
"requires_python": "",
"maintainer_email": "",
"keywords": "",
"author": "Habana Labs, Ltd. an Intel Company",
"author_email": "support@habana.ai",
"download_url": "",
"platform": null,
"description": "# Habana TensorFlow Python package\n\n`habana-tensorflow` is a package designed for easy integration of Gaudi bridge to Tensorflow topologies.\nMain entry point (Python import) is `habana_frameworks.tensorflow` module that contains all the necessary functions to work with Gaudi.\n\n## Structure\n\nProperly built wheel contains:\n- `habana_frameworks` python namespace (with all the folder structure inside)\n- libraries for all supported versions of TF (see `habana_frameworks.tensorflow.sysconfig._supported_tf_versions`) located in `habana_frameworks/tensorflow/tf<version_with_underscores>/lib/habanalabs/`\n- proper licensing\n\nWhen loading `habana_tensorflow`, there is sanity check for currently installed TF version in the system, to match libraries found in the module.\n\n## TensorFlow package (habana_frameworks.tensorflow)\n\nMost important module inside is ``library_loader`` and function `load_habana_module()`, that is exposed directly after import.\nIt initializes the module properly.\n\nExample:\n\n```python\nimport habana_frameworks.tensorflow as htf\nhtf.load_habana_module()\n```\n\n### List of modules inside ``habana_frameworks.tensorflow``\n\nAll modules are described in [official docs](https://docs.habana.ai/en/latest/Tensorflow_User_Guide/Tensorflow_User_Guide.html#python-package-habana-frameworks-tensorflow).\n",
"bugtrack_url": null,
"license": "See LICENSE.txt and third-party-programs.txt",
"summary": "This package contains Habana\u00ae TensorFlow bridge that allows to work with Habana\u00ae Gaudi\u00ae",
"version": "1.14.0.493",
"project_urls": {
"Homepage": "https://habana.ai/"
},
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "2e2acfddd5a892a042332bb8700915e60b00acf6e994dcadb1be7e28a40bc5dd",
"md5": "2b9bae6fd0351eb6a89a2f464852d742",
"sha256": "546a4ad5c481ef1dd19bf1d6103a4e01a6203a72cd54b2b0feddf55d2207e614"
},
"downloads": -1,
"filename": "habana_tensorflow-1.14.0.493-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "2b9bae6fd0351eb6a89a2f464852d742",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": null,
"size": 6712729,
"upload_time": "2024-01-23T11:56:44",
"upload_time_iso_8601": "2024-01-23T11:56:44.993568Z",
"url": "https://files.pythonhosted.org/packages/2e/2a/cfddd5a892a042332bb8700915e60b00acf6e994dcadb1be7e28a40bc5dd/habana_tensorflow-1.14.0.493-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-01-23 11:56:44",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "habana-tensorflow"
}
JSON