Quick start

Get from zero to a running local ML Studio in a few minutes.

1. Install

pip install "oneopen-ml-studio[yolo]"

2. Initialize data directory

oneopen init

Creates ~/.oneopen (or ONEOPEN_DATA_DIR) with SQLite and folders for projects, exports, and models.

3. Start the server

oneopen start

Open http://127.0.0.1:8765.

(oneopen serve is the same command.)

Important

Default bind is 127.0.0.1 — local-first, no built-in authentication. Team/server auth is on the roadmap.

Useful options

oneopen start --port 9000
oneopen start --host 127.0.0.1 --reload

4. Create a project

  1. New project (or /projects/new).

  2. Name + type (object detection is the primary path today).

  3. Optional custom location; blank → ~/.oneopen/projects/<id>/.

5. Add classes and images

  1. Add at least one class.

  2. Upload images or import a YOLO folder/zip.

6. Annotate

  1. Open Annotate.

  2. Draw boxes or polygons.

  3. Optionally run Label Assist (YOLO), then correct.

7. Version, train, or export

  1. Assign splits.

  2. Create a dataset version.

  3. Export and/or Train locally.

Typical first workflow

Create project → Add classes → Upload images
        ↓
   Annotate (manual + Label Assist)
        ↓
 Assign splits → Generate version
        ↓
   Export YOLO  and/or  Train locally

Next steps