TensorRT Performance: How to extract the metric Efficiency

Hi all, I am running the samples included in TensorRT and need to know how to extract the metric Efficiency (images/sec/Watt) as shown in this article:
https://news.developer.nvidia.com/nvidia-ai-inference-performance-milestones-delivering-leading-throughput-latency-and-efficiency/

this is something we are working on. stay tuned.