Export a dataset. Keep a project.

Export selected data as CSV

CSV exports are for working with data outside the editor. They are not a replacement for saving a complete editable project.

1. Open the folder containing your reviewed, labeled segments.
2. Choose Export CSV and review the dataset export options.
3. Choose the included data and metadata deliberately. Leave out participant details you do not need.
4. Choose the destination and export. Open the result in your analysis tool and check the columns, units, labels, and sample count before using it.

Prepare data for Create ML

Use Export Create ML from a folder when you need the app’s activity-data export format. Check the export report for skipped items and review the output before training.

A successful export does not establish dataset quality or model accuracy. Use representative recordings, consistent labels, and an appropriate evaluation strategy for your project.

Save a project you can reopen

In the Mac app’s Settings → Export, you can change the default project name, whether the date and time are appended, and the project format. .watchmotion and .zip contain the same ZIP-based project. CSV datasets stay .csv; metadata stays .json.

Keep exported projects and datasets private when they include participant information or sensitive notes. A file saved to a cloud-synced folder is also subject to that storage provider’s settings.

1. Choose Export Project in the workspace toolbar.
2. Choose a file name and destination. The default project extension is .watchmotion.
3. Keep the project file somewhere you control. Use Open Project to continue working with it later.