Installation¶

Use Python 3.10 or newer. A virtual environment keeps the package and its dependencies separate from other projects:

python -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install ml-pipes-ultralytics

On Windows, activate the environment with .venv\Scripts\activate instead. The package installs ml-pipes-core, ml-pipes-vision, and Ultralytics, along with their dependencies, including NumPy, OpenCV, and PyTorch. The tutorials do not require a separate vision-operator installation.

If working from a clone of this repository, install the checkout from its root directory instead of the published package:

python -m pip install -e .

Model checkpoints such as yolo26n.pt may be downloaded when a model operator is constructed. Sample-image downloads and environment/model setup should be completed before measuring inference performance.

For GPU acceleration, use a PyTorch build appropriate for your hardware; see the PyTorch installation instructions. Configure the model operator's device option when selecting a particular device. The guides also work on CPU; performance depends on the machine.

Ultralytics software and model weights have separate licensing terms. Read the licensing note before using or redistributing them.