Install YOLO¶

This guide will help you set up YOLO on your machine. We recommend starting with Git & GitHub for more flexible customization. If you only need inference or simple customization, install via pip.

Torch Requirements¶

Git & GitHub¶

Clone the repository:

git clone https://github.com/shreyaskamathkm/yolo.git

Or download directly via the ZIP archive.

Install all dependencies:

make setup

This creates a .venv virtual environment (if one doesn't exist), installs all dependencies (requirements-dev.txt + editable package install), and configures pre-commit hooks. You can override defaults:

make setup VENV=myenv PYTHON=python3.11

For ONNX or TensorRT setup, see Inference & Deployment.

PyPI (pip install)¶

Note

Due to the yolo name already being occupied on PyPI, we provide installation via GitHub. Ensure git and pip are available in your shell.

pip install git+https://github.com/shreyaskamathkm/yolo.git

Docker¶

docker pull shreyaskamathkm/yolo
docker run --gpus all -it shreyaskamathkm/yolo

Requires the NVIDIA Container Toolkit.

Conda¶

Coming soon.