Roboflow Custom Metadata¶
Version v1¶
Block allows users to add custom metadata to each inference result in Roboflow Model Monitoring dashboard.
This is useful for adding information specific to your use case. For example, if you want to be able to filter inferences by a specific label such as location, you can attach those labels to each inference with this block.
For more information on Model Monitoring at Roboflow, see https://docs.roboflow.com/deploy/model-monitoring.
Type identifier¶
Use the following identifier in step "type" field: roboflow_core/roboflow_custom_metadata@v1to add the block as
as step in your workflow.
Properties¶
| Name | Type | Description | Refs |
|---|---|---|---|
name |
str |
The unique name of this step.. | ❌ |
field_value |
str |
This is the name of the metadata field you are creating. | ✅ |
field_name |
str |
Name of the field to be updated in Roboflow Customer Metadata. | ❌ |
fire_and_forget |
bool |
Boolean flag dictating if sink is supposed to be executed in the background, not waiting on status of registration before end of workflow run. Use True if best-effort registration is needed, use False while debugging and if error handling is needed. |
✅ |
The Refs column marks possibility to parametrise the property with dynamic values available
in workflow runtime. See Bindings for more info.
Available Connections¶
Check what blocks you can connect to Roboflow Custom Metadata in version v1.
- inputs:
Detection Offset,Detections Consensus,Template Matching,OCR Model,Detections Stitch,Multi-Label Classification Model,Detections Classes Replacement,Segment Anything 2 Model,Roboflow Dataset Upload,LMM,Object Detection Model,VLM as Detector,Roboflow Custom Metadata,VLM as Classifier,Perspective Correction,OpenAI,Clip Comparison,OpenAI,CogVLM,Keypoint Detection Model,Anthropic Claude,Detections Filter,Instance Segmentation Model,LMM For Classification,Google Gemini,YOLO-World Model,Detections Transformation,Roboflow Dataset Upload,Single-Label Classification Model - outputs:
Roboflow Custom Metadata
The available connections depend on its binding kinds. Check what binding kinds
Roboflow Custom Metadata in version v1 has.
Bindings
-
input
predictions(Union[keypoint_detection_prediction,object_detection_prediction,instance_segmentation_prediction,classification_prediction]): Reference data to extract property from.field_value(string): This is the name of the metadata field you are creating.fire_and_forget(boolean): Boolean flag dictating if sink is supposed to be executed in the background, not waiting on status of registration before end of workflow run. UseTrueif best-effort registration is needed, useFalsewhile debugging and if error handling is needed.
-
output
Example JSON definition of step Roboflow Custom Metadata in version v1
{
"name": "<your_step_name_here>",
"type": "roboflow_core/roboflow_custom_metadata@v1",
"predictions": "$steps.my_step.predictions",
"field_value": "toronto",
"field_name": "The name of the value of the field",
"fire_and_forget": true
}