Installation

OmniCloudMask requires Python 3.9 or higher.

Install from PyPI

pip install omnicloudmask

Or using uv:

uv add omnicloudmask

Install from conda-forge

conda install conda-forge::omnicloudmask

Install from source

pip install git+https://github.com/DPIRD-DMA/OmniCloudMask.git

Docker

Prerequisites:

• Docker installed

• For GPU support: NVIDIA Container Toolkit

Build the Docker image:

git clone https://github.com/DPIRD-DMA/OmniCloudMask
cd OmniCloudMask
docker build -f docker/Dockerfile -t omnicloudmask:local .

Run with GPU support:

docker run --gpus all -p 127.0.0.1:8888:8888 omnicloudmask:local

Or CPU only:

docker run -p 127.0.0.1:8888:8888 omnicloudmask:local

Mount a data directory:

docker run --gpus all -p 127.0.0.1:8888:8888 -v /path/to/data:/workspace/data omnicloudmask:local

Access Jupyter Lab at http://localhost:8888 (no password required).

Legacy Model Support

Model versions 1.0 through 3.0 were built with fastai. To use these older models, install the legacy extra:

pip install omnicloudmask[legacy]

Or with uv:

uv add omnicloudmask --extra legacy

Or with conda:

conda install conda-forge::omnicloudmask conda-forge::fastai

See the model changelog for differences between model versions.