Skip to main content
Version: 0.4.x

Denkflow Integration Guide

Alpha Status

This software is currently in alpha status. This means that:

  • It is suitable for testing environments but caution is advised for production use
  • Some bugs may still be present
  • The API is stabilizing but minor changes may still occur
  • Most core features are implemented and working correctly
Planned Features

The following features are planned for upcoming releases:

  • Support for all model types (classification, segmentation, rotated object detection, etc.)
  • USB Dongle Support
  • C API for integration with C/C++ applications
  • C# API for .NET applications
  • Public Rust API for high-performance integrations
  • Output Modules
  • etc.

Overview

Denkflow is DENKweit's solution for a complete end-to-end integration, from training AI models on our Vision AI Hub to executing them in your production environment.

Currently, object detection and ocr are implemented for CPU, NVIDIA GPU and Jetson Orin devices.

Quick Start​

from denkflow import Pipeline, ImageTensor

pipeline = Pipeline.from_denkflow(model_file, pat=pat) # pat = personal access token

# --- Initialization ---
pipeline.initialize()

# --- Subscribe to Outputs ---
receiver = pipeline.subscribe_bounding_box_tensor("bounding_box_filter_node/filtered_bounding_boxes")

# --- Publish Image ---
pipeline.publish_image_tensor("camera/image", ImageTensor.from_file("path/to/your/image.jpg"))

# --- Run Pipeline ---
pipeline.run()

# --- Receive and Process Results ---
objects = receiver.receive().to_objects(0.5)

print(f"\nDetected {len(objects)} objects:")
for obj in objects:
    print(f"- Class: {obj.class_label.name}, Confidence: {obj.confidence:.2f}")
    print(f"  BBox: ({obj.x1}, {obj.y1}), ({obj.x2}, {obj.y2})")