tequ-setup-triton-inference-server
Configure NVIDIA Triton Inference Server on different platforms. Deploy object detection model in Tensorflow SavedModel format to server. Send images to server for inference with Node-RED. Triton Inference Server HTTP API is used for inference.