Installation

1. Requirements
2. Install from PyPI
3. Cloud backends (optional extras)
4. Develop from source
5. Next

Requirements

• Python 3.11+
• Works on Linux, macOS, and Windows.

Install from PyPI

pip install visionpack

This installs the vp command (and a visionpack alias). Verify it:

vp --version
vp --help

Prefer an isolated install? pipx keeps the CLI off your global environment: pipx install visionpack.

Cloud backends (optional extras)

Reading from or writing to object stores pulls in fsspec plus the provider library. They’re optional — the core never imports them, so a local install stays lean. Install only what you need:

pip install "visionpack[s3]"      # Amazon S3   (s3fs)
pip install "visionpack[gcs]"     # Google Cloud Storage (gcsfs)
pip install "visionpack[azure]"   # Azure Blob  (adlfs)

You can combine them: pip install "visionpack[s3,gcs]". See Cloud Sync for declaring remote sources and a cloud target.

Develop from source

VisionPack uses uv. From the repo root:

git clone https://github.com/CaioWing/VisionPack
cd VisionPack
uv sync                 # create the venv and install dependencies
uv run vp --help        # run the CLI

Run the test suite:

uv run python -m unittest discover -s tests

You can also invoke the module directly without the console script:

uv run python -m visionpack --help

Next

• Quickstart — build a dataset end to end.
• CLI Guide — the full command reference.