Data, Project and Export
The main project stores the task being edited. Images, method results, and representations for other tools are stored next to it as separate files.
What is included in the project
- Format version, program mode and task settings.
- Elements, features, regions, relationships, groups and annotations.
- Links to source images and attached computer vision results.
- Information you need to check the structure and associated files when you open them.
Files next to the project
| Material | Purpose |
|---|---|
| Source images and frames | Visual materials on which the markup was created. |
| Results of methods | Masks, outlines, detections, recognized text, metrics and execution details. |
| Code view | Selected task data for Screph Code or a supported external tool. It does not replace the main project. |
| Automation script | A separate code file that uses the main project when explicitly running Automation Runtime. |
Computer vision results
Previewing and temporarily running a method does not change the underlying markup. The result can be attached to the project, and the proposed change to an element or characteristic is applied only after verification and confirmation.
When you save a project, only the materials that are associated with the project by the selected workflow are copied. The remaining temporary results remain in the local data directory and are cleaned up separately.
Transfer and backup
- Save the project and wait until the files are written.
- Copy the entire project directory along with images and attached results.
- Open the copy from the new location and check for warnings about missing files.
- If your project uses a separate script or working code folder, save them separately.
Coding and Execution
- Screph Code gets the selected data view and can modify the files in the working directory.
- Automation Runtime opens the main project and a separately prepared script.
- Passing data does not run code or dispatch input actions automatically.
Before handing over the project
- Remove unnecessary images, logs, keys, personal information, and experiment results.
- Check the licenses of the source materials and models.
- Open a copy of the project and make sure all the necessary files are available.