Node-red support for YOLO kubeflow pipeline

[kai910120]

You only look once (YOLO) is a state-of-the-art, real-time object detection system. On a Pascal Titan X it processes images at 30 FPS and has a mAP of 57.9% on COCO test-dev. Although it is a powerfull tool for AI applications, many coding procedures including proper library infusion and grammer editting etc. are needed. It is a nice thing to have an automation process to package all these operations into a kubeflow pipeline. Hence we propose a YOLO kubeflow pipeline as a convenient tool for major ML operations for object detection. Further, it takes many laborious procedures to build a practical YOLO pipelines, e.g. docker image editing and YAML file compilation etc. A graphical UI can provide a rapid deployment medium for ML developer especially YOLO, professional/non professional alike. Here we propose a Node-red frontend which can exercise an YOLO operation without starting from scratch. Node-red is a graphical user interface that features drag and drop operations on its pallet panel for easy customization of YOLO functionality. Please let us know if such contribution is proper for this site. We will be doing PR soon on this site.