How to measure accuracy or performance metrics of YOLO model being used in sample object detector tracker?

Please provide the following info (tick the boxes after creating this topic):
Software Version
DRIVE OS 6.0.10.0
[1] DRIVE OS 6.0.8.1
DRIVE OS 6.0.6
DRIVE OS 6.0.5
DRIVE OS 6.0.4 (rev. 1)
DRIVE OS 6.0.4 SDK
other

Target Operating System
[1] Linux
QNX
other

Hardware Platform
DRIVE AGX Orin Developer Kit (940-63710-0010-300)
DRIVE AGX Orin Developer Kit (940-63710-0010-200)
DRIVE AGX Orin Developer Kit (940-63710-0010-100)
DRIVE AGX Orin Developer Kit (940-63710-0010-D00)
DRIVE AGX Orin Developer Kit (940-63710-0010-C00)
[1] DRIVE AGX Orin Developer Kit (not sure its number)
other

SDK Manager Version
2.1.0
[1] other

Host Machine Version
native Ubuntu Linux 20.04 Host installed with SDK Manager
native Ubuntu Linux 20.04 Host installed with DRIVE OS Docker Containers
native Ubuntu Linux 18.04 Host installed with DRIVE OS Docker Containers
other

Issue Description
Hi team, I executed a sample object detector tracker code. I observed YOLO has been used to detect objects. How can I measure the performance metrics of YOLO model to know how accurately a model is detecting objects? Is this data provided?

Dear @akshay.tupkar ,
We don’t provide any accuracy scripts for DNN model. You can try with any public data like COCO and feed them to TensorRT model and store output to check accuracy later.

Dear @SivaRamaKrishnaNV
Do I need to implement this in the same existing code?

Dear @SivaRamaKrishnaNV
May I know the yolo variant used in sample object detector tracker on Drive OS 6.0.8.1?

Yolo v3

If you asking like perf metrics like mAP, You can verify it inside a TRT sample like in TensorRT/samples/python/yolov3_onnx at release/8.6 · NVIDIA/TensorRT · GitHub with input images from dataset.

Thanks for sharing this @SivaRamaKrishnaNV
I tried this, generated onnx model. But while generating tensorrt engine from onnx model, it throws some errors. Because recent versions of tensorrt expects a onnx model to be defined explicitly with batch dimension.