Metadata-Version: 2.3
Name: pixano-inference
Version: 0.3.1
Summary: Inference models for Pixano, data-centric AI building blocks for computer vision applications
Project-URL: Documentation, https://github.com/pixano/pixano-inference#readme
Project-URL: Issues, https://github.com/pixano/pixano-inference/issues
Project-URL: Source, https://github.com/pixano/pixano-inference
Author-email: Pixano Developers <pixano@cea.fr>
License-Expression: CECILL-C
License-File: LICENSE
Keywords: computer vision,data annotation,data visualization,deep learning,machine learning
Classifier: Development Status :: 4 - Beta
Classifier: License :: CeCILL-C Free Software License Agreement (CECILL-C)
Classifier: Programming Language :: Python :: 3.10
Requires-Python: ~=3.10
Requires-Dist: gitpython>=3.1.30
Requires-Dist: matplotlib>=3.3
Requires-Dist: pixano~=0.5.0
Requires-Dist: psutil>=5.9.5
Requires-Dist: pyyaml>=5.3.1
Requires-Dist: requests>=2.23.0
Requires-Dist: scipy>=1.4.1
Requires-Dist: seaborn>=0.11.0
Requires-Dist: tensorflow-hub~=0.15.0
Requires-Dist: tensorflow~=2.15.0
Requires-Dist: thop>=0.1.1
Requires-Dist: timm>=0.9.12
Requires-Dist: torchaudio~=2.2.0
Requires-Dist: torchvision~=0.17.0
Requires-Dist: torch~=2.2.0
Requires-Dist: transformers>=4.33.0
Provides-Extra: documentation
Requires-Dist: mkdocs-gen-files~=0.5.0; extra == 'documentation'
Requires-Dist: mkdocs-literate-nav~=0.6.0; extra == 'documentation'
Requires-Dist: mkdocs-material~=9.4.0; extra == 'documentation'
Requires-Dist: mkdocs-redirects~=1.2.1; extra == 'documentation'
Requires-Dist: mkdocstrings-python~=1.7.0; extra == 'documentation'
Description-Content-Type: text/markdown

<div align="center">

<img src="https://raw.githubusercontent.com/pixano/pixano-inference/main/docs/assets/pixano_wide.png" alt="Pixano" height="100"/>

<br/>

**Inference models for Pixano**

**_Under active development, subject to API change_**

[![GitHub version](https://img.shields.io/github/v/release/pixano/pixano-inference?label=release&logo=github)](https://github.com/pixano/pixano-inference/releases)
[![PyPI version](https://img.shields.io/pypi/v/pixano-inference?color=blue&label=release&logo=pypi&logoColor=white)](https://pypi.org/project/pixano-inference/)
[![Documentation](https://img.shields.io/website?url=https%3A%2F%2Fpixano.github.io%2F&up_message=online&down_message=offline&label=docs)](https://pixano.github.io)
[![Python version](https://img.shields.io/pypi/pyversions/pixano?color=important&logo=python&logoColor=white)](https://www.python.org/downloads/)
[![License](https://img.shields.io/badge/license-CeCILL--C-blue.svg)](LICENSE)

</div>

<hr />

<a href="https://github.com/pixano/pixano" target="_blank">**Pixano**</a> is an open-source tool by CEA List for exploring and annotating your dataset using AI features like **smart segmentation** and **semantic search**.

**Pixano Inference** provides the AI models like _SAM_ and _CLIP_ that power those features, as well as a PyTorch and TensorFlow models for pre-annotating your datasets.

# Getting started

## Installing Pixano Inference

As Pixano and Pixano Inference require specific versions for their dependencies, we recommend creating a new Python virtual environment to install them.

For example, with <a href="https://conda.io/projects/conda/en/latest/user-guide/install/index.html" target="_blank">conda</a>:

```shell
conda create -n pixano_env python=3.10
conda activate pixano_env
```

Then, you can install Pixano Inference inside that environment with pip:

```shell
pip install pixano-inference
```

As it is a requirement of Pixano Inference, Pixano will automatically be downloaded if it is not already installed.

## Using the models

Please refer to <a href="https://github.com/pixano/pixano/tree/main/notebooks/models" target="_blank">these notebooks</a> for information on how to use the models provided by Pixano Inference.

# Contributing

Please refer to the [CONTRIBUTING.md](CONTRIBUTING.md) for information on running Pixano locally and guidelines on how to publish your contributions.

# License

Pixano Inference is licensed under the [CeCILL-C license](LICENSE).
