The Settings tab allows you to manage various configurations such as:
If you have the Enterprise edition, you can also configure connections to Gravio Coordinator and blockchain features.
Here, you can view various information about the connected HubKit. This information is important for the Asteria team to know which version you are using when reporting bugs.
At the bottom of the screen, you can configure email server parameters for notifications. These notifications provide information about disk space shortages and HubKit issues.
Gravio 4.5+ Feature: If no regular communication (approximately every 60 minutes) is received from a Zigbee device and this state continues for a certain period (90 minutes), a notification will be sent to the email address registered in the notification settings indicating that communication from the device has been lost. For CO2 sensors, notifications are sent if no signal is received for 1 minute.
You can backup device configuration settings to a Gravio cloud account (either hosted by Asteria or your own local Gravio Coordinator). From there, you can restore these settings to your Gravio account’s home folder.
These are reusable settings that are not included in actions or backups. This includes authentication tokens, passwords, login details, secret keys, and email server settings. This allows you to reuse them in actions as needed.
Feature packages are configuration sets including areas, layers, triggers, and actions that can be exported as Zip files and imported into cloud accounts. If you have the Gravio Enterprise edition, you can also import them into the Coordinator for distribution to edge nodes.
Note: It is recommended to put all sensitive data such as tokens, usernames, and third-party service passwords into base property profiles (see above). This ensures that these details are not exported in the package if the appropriate checkbox is checked during the export process.
This section allows you to deploy available image inference models to your local HubKit. There are two ways to deploy computer vision models:
There are also several standard models provided by Gravio Cloud, including pre-trained models for people counting.
To upload a new computer vision model locally via Gravio Studio:
.tflite file packageAlternatively, you can click the “Import” button to import a previously exported zip file.
For TensorFlow, prepare the following files:
For TensorFlow Lite, prepare the following files:
Form settings explanation:
| Field | Description |
|---|---|
| Model Package Name | Use alphanumeric characters to describe the package. This name will be displayed as a “sensor”. |
| Method | Choose “Count” to output numbers or “Grouping” to return strings (cannot be specified if “Image Classification” is selected) |
| Output Format | Choose “JSON” to output detailed information or “Value” to output only values (for image classification, choose “Name”) |
| Include Detection Values | Select this checkbox to include detected values |
| Confidence Threshold | Set between 0.0 and 1.0 to define the threshold level at which the “sensor” is triggered |
Disk Management displays the health status of the disk and allows you to configure various parameters for optimal disk usage.
You can view information about disk usage, log files, and media data. You can also configure cleanup settings and specify retention periods for media data and log files.