Model Monitoring Inference Aggregator

Class: ModelMonitoringInferenceAggregatorBlockV1

Source: inference.core.workflows.core_steps.sinks.roboflow.model_monitoring_inference_aggregator.v1.ModelMonitoringInferenceAggregatorBlockV1

This block 📊 transforms inference data reporting to a whole new level by periodically aggregating and sending a curated sample of predictions to Roboflow Model Monitoring.

✨ Key Features

• Effortless Aggregation: Collects and organizes predictions in-memory, ensuring only the most relevant and confident predictions are reported.

• Customizable Reporting Intervals: Choose how frequently (in seconds) data should be sent—ensuring optimal balance between granularity and resource efficiency.

• Debug-Friendly Mode: Fine-tune operations by enabling or disabling asynchronous background execution.

🔍 Why Use This Block?

This block is a game-changer for projects relying on video processing in Workflows. With its aggregation process, it identifies the most confident predictions across classes and sends them at regular intervals in small messages to Roboflow backend - ensuring that video processing performance is impacted to the least extent.

Perfect for:

• Monitoring production line performance in real-time 🏭.

• Debugging and validating your model’s performance over time ⏱️.

• Providing actionable insights from inference workflows with minimal overhead 🔧.

🚨 Limitations

• The block is should not be relied on when running Workflow in inference server or via HTTP request to Roboflow hosted platform, as the internal state is not persisted in a memory that would be accessible for all requests to the server, causing aggregation to only have a scope of single request. We will solve that problem in future releases if proven to be serious limitation for clients.

• This block do not have ability to separate aggregations for multiple videos processed by InferencePipeline - effectively aggregating data for all video feeds connected to single process running InferencePipeline.

Type identifier

Use the following identifier in step "type" field: roboflow_core/model_monitoring_inference_aggregator@v1to add the block as as step in your workflow.

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
frequency	int	Frequency of reporting (in seconds). For example, if 5 is provided, the block will report an aggregated sample of predictions every 5 seconds..	✅
unique_aggregator_key	str	Unique key used internally to track the session of inference results reporting. Must be unique for each step in your Workflow..	❌
fire_and_forget	bool	Boolean flag to run the block asynchronously (True) for faster workflows or synchronously (False) for debugging and error handling..	✅

The Refs column marks possibility to parametrise the property with dynamic values available in workflow runtime. See Bindings for more info.

Available Connections

Compatible Blocks

Check what blocks you can connect to Model Monitoring Inference Aggregator in version v1.

• inputs: Detections Filter, Detections Stitch, Detections Classes Replacement, Path Deviation, Time in Zone, Template Matching, Model Monitoring Inference Aggregator, Roboflow Dataset Upload, Overlap Filter, Gaze Detection, Single-Label Classification Model, Slack Notification, Perspective Correction, Roboflow Dataset Upload, Anthropic Claude, OpenAI, Florence-2 Model, Object Detection Model, Keypoint Detection Model, Llama 3.2 Vision, Bounding Rectangle, Dynamic Crop, OCR Model, EasyOCR, Velocity, Keypoint Detection Model, Detection Offset, Identify Changes, Florence-2 Model, Moondream2, SIFT Comparison, Byte Tracker, Dynamic Zone, Stitch OCR Detections, Detections Transformation, Multi-Label Classification Model, Byte Tracker, Detections Combine, YOLO-World Model, Clip Comparison, VLM as Detector, Google Vision OCR, OpenAI, Line Counter, Webhook Sink, Instance Segmentation Model, CogVLM, Time in Zone, Twilio SMS Notification, Detections Consensus, SIFT Comparison, Multi-Label Classification Model, LMM, Time in Zone, JSON Parser, LMM For Classification, Detections Stabilizer, Instance Segmentation Model, Segment Anything 2 Model, VLM as Classifier, VLM as Detector, Email Notification, Local File Sink, Roboflow Custom Metadata, Google Gemini, Object Detection Model, VLM as Classifier, PTZ Tracking (ONVIF).md), CSV Formatter, Path Deviation, Detections Merge, OpenAI, Byte Tracker, Identify Outliers, Single-Label Classification Model
• outputs: CLIP Embedding Model, Detections Stitch, Circle Visualization, Time in Zone, Model Monitoring Inference Aggregator, QR Code Generator, Dot Visualization, Gaze Detection, Single-Label Classification Model, Slack Notification, Blur Visualization, Florence-2 Model, Object Detection Model, Keypoint Detection Model, Llama 3.2 Vision, Crop Visualization, Image Threshold, Triangle Visualization, Reference Path Visualization, Model Comparison Visualization, Cache Set, Corner Visualization, Image Blur, Size Measurement, SIFT Comparison, Bounding Box Visualization, Dynamic Zone, Stitch OCR Detections, Multi-Label Classification Model, YOLO-World Model, Distance Measurement, Icon Visualization, Stability AI Inpainting, Google Vision OCR, Polygon Zone Visualization, Webhook Sink, CogVLM, Time in Zone, Cache Get, Stability AI Outpainting, Classification Label Visualization, Multi-Label Classification Model, Morphological Transformation, Instance Segmentation Model, Segment Anything 2 Model, Local File Sink, Roboflow Custom Metadata, PTZ Tracking (ONVIF).md), Pixelate Visualization, Pixel Color Count, Path Deviation, Single-Label Classification Model, Detections Classes Replacement, Path Deviation, Template Matching, Roboflow Dataset Upload, Perception Encoder Embedding Model, Perspective Correction, Roboflow Dataset Upload, Anthropic Claude, Background Color Visualization, OpenAI, Dynamic Crop, Trace Visualization, Keypoint Detection Model, Polygon Visualization, Florence-2 Model, Moondream2, Line Counter, Keypoint Visualization, Clip Comparison, Line Counter Visualization, OpenAI, Line Counter, Instance Segmentation Model, Mask Visualization, Twilio SMS Notification, Detections Consensus, LMM, Image Preprocessing, Time in Zone, Color Visualization, LMM For Classification, Ellipse Visualization, Stability AI Image Generation, Email Notification, Halo Visualization, Google Gemini, Object Detection Model, Label Visualization, OpenAI, Contrast Equalization

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds Model Monitoring Inference Aggregator in version v1 has.

Bindings

• input

  • predictions (Union[object_detection_prediction, instance_segmentation_prediction, keypoint_detection_prediction, classification_prediction]): Model predictions to report to Roboflow Model Monitoring..
  • model_id (roboflow_model_id): Model ID to report to Roboflow Model Monitoring..
  • frequency (string): Frequency of reporting (in seconds). For example, if 5 is provided, the block will report an aggregated sample of predictions every 5 seconds..
  • fire_and_forget (boolean): Boolean flag to run the block asynchronously (True) for faster workflows or synchronously (False) for debugging and error handling..

• output

  • error_status (boolean): Boolean flag.
  • message (string): String value.

Example JSON definition of step Model Monitoring Inference Aggregator in version v1

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/model_monitoring_inference_aggregator@v1",
    "predictions": "$steps.my_step.predictions",
    "model_id": "my_project/3",
    "frequency": "3",
    "unique_aggregator_key": "session-1v73kdhfse",
    "fire_and_forget": true
}