Metadata-Version: 2.4
Name: romiseg
Version: 0.3.0
Summary: Image multiclass semantic segmentation using CNN models trained on virtual plant images (PyTorch).
Author-email: Alienor Lahlou <alienor.lahlou@espci.org>, Peter Hanappe <peter.hanappe@sony.com>
Maintainer-email: Jonathan Legrand <jonathan.legrand@ens-lyon.fr>
License: LGPL-3.0-or-later
Project-URL: homepage, https://romi-project.eu/
Project-URL: documentation, https://docs.romi-project.eu/plant_imager/
Project-URL: source, https://github.com/romi/romiseg
Project-URL: issues, https://github.com/romi/romiseg/issues
Keywords: Robotics for Microfarms,ROMI,U-Net,Deep CNN,Virtual Plant,semantic segmentation,PyTorch
Classifier: Programming Language :: Python :: 3
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Science/Research
Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Operating System :: OS Independent
Classifier: Natural Language :: English
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: appdirs
Requires-Dist: future
Requires-Dist: Pillow
Requires-Dist: requests
Requires-Dist: torch==2.13.0
Requires-Dist: torchvision<0.29,>=0.28
Requires-Dist: tqdm
Requires-Dist: plantdb.commons
Provides-Extra: train
Requires-Dist: labelme; extra == "train"
Requires-Dist: matplotlib; extra == "train"
Requires-Dist: pyqt5==5.14; extra == "train"
Requires-Dist: tensorboard; extra == "train"
Provides-Extra: doc
Requires-Dist: mkdocs; extra == "doc"
Requires-Dist: mkdocs-material; extra == "doc"
Requires-Dist: mkdocstrings[python]; extra == "doc"
Requires-Dist: mkdocstrings-python<2; extra == "doc"
Requires-Dist: pymdown-extensions; extra == "doc"
Requires-Dist: markdown-exec[ansi]; extra == "doc"
Requires-Dist: mkdocs-gen-files; extra == "doc"
Requires-Dist: mkdocs-literate-nav; extra == "doc"
Requires-Dist: mkdocs-section-index; extra == "doc"
Dynamic: license-file

# [![ROMI_logo](docs/assets/images/ROMI_logo_green_25.svg)](https://romi-project.eu) / romiseg

[![Licence](https://img.shields.io/github/license/romi/romiseg?color=lightgray)](https://www.gnu.org/licenses/lgpl-3.0.en.html)
[![Python Version](https://img.shields.io/python/required-version-toml?tomlFilePath=https%3A%2F%2Fraw.githubusercontent.com%2Fromi%2Fromiseg%2Frefs%2Fheads%2Fdev%2Fpyproject.toml&logo=python&logoColor=white)]()
[![PyPI - Version](https://img.shields.io/pypi/v/romiseg?logo=pypi&logoColor=white)](https://pypi.org/project/romiseg/)
[![Conda - Version](https://img.shields.io/conda/vn/romi-eu/romiseg?logo=anaconda&logoColor=white&label=romi-eu&color=%2344A833)](https://anaconda.org/romi-eu/romiseg)
[![GitHub branch check runs](https://img.shields.io/github/check-runs/romi/romiseg/dev)](https://github.com/romi/romiseg)

## Overview

This package contains trained CNN models and methods dedicated to performs semantic segmentation of plant images.

## Installation

### Clone the sources

```shell
git clone https://github.com/romi/romiseg
```

### Optional - Create a conda environment

```shell
conda create -n romiseg python==3.10
```

### Install PyTorch & dependencies

```shell
python -m pip install 'torch>=2.0.0' 'torchvision>=0.15.0' --extra-index-url 'https://download.pytorch.org/whl/cu118'
```

We install it for CUDA 11.8, which is compatible with the latest PyTorch 2.x versions.

### Install `romiseg`

```shell
cd romiseg
python -m pip install -e .
```

## Usage and Documentation

- Segmentation2D module: outputs, prerequisites, and details are documented at docs/segmentation2d.md
- Annotation and fine‑tuning: step‑by‑step guide at docs/annotation-finetuning.md

For the full documentation website, see https://romi.github.io/romiseg/
