Metadata-Version: 2.1
Name: finegrained
Version: 0.1.1
Summary: Quickly prepare data, train new models and export to production from a command line
License: MIT
Author: Rustem Galiullin
Author-email: rustemgal@gmail.com
Requires-Python: >=3.9,<4.0
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.9
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3
Requires-Dist: fiftyone (>=0.21.6,<0.22.0)
Requires-Dist: fire (>=0.5.0,<0.6.0)
Requires-Dist: numpy (>=1.22.4,<2.0.0)
Requires-Dist: pillow (>=9.4.0,<10.0.0)
Requires-Dist: pyyaml (>=6.0,<7.0)
Requires-Dist: scikit-learn (>=1.2.2,<2.0.0)
Requires-Dist: tqdm (>=4.65.0,<5.0.0)
Description-Content-Type: text/markdown

# Fine-grained image recognition

A package and command-line tool to quickly get from raw data to a production model.
The main target is to train an object detector and a image classfier / similarity search
model to detect and recognize fine-grained objects on images and video.

This package is built upon an open-source computer vision data management tool
[FiftyOne](https://docs.voxel51.com/index.html). It also provides an integration
with a data annotation tool [CVAT](https://www.cvat.ai/) and a model-training framework
[Lightning Flash](https://lightning-flash.readthedocs.io/) that are 

## Structure

The package is structured as follows:

- **data**: contains the code to prepare data for training
- **models**: contains the code to train, evaluate and export models

## Development installation

This package is managed with [Poetry](https://python-poetry.org/).
To install the package in development mode, run the following command:

```bash
poetry install
pre-commit install
```

