INT8 Calibration has error on Jetson AGX Orin Developer Kit

[lakshanthad]

Hardware: Jetson AGX Orin Developer Kit

Software: JetPack 5.0.1 DP

What works: Inference works well with FP32

Issue: Inference does not work with INT8. The following output log can be seen after "deepstream-app -c deepstream_app_config.txt"

Output YOLO blob names: 
yolo_93
yolo_96
yolo_99

Total number of YOLO layers: 260

Building YOLO network complete
Building the TensorRT Engine

NOTE: letter_box is set in cfg file, make sure to set maintain-aspect-ratio=1 in config_infer file to get better accuracy


File does not exist: /mnt/sd/workspace/customer/seeed/DeepStream-Yolo/calib.table
ERROR: [TRT]: 1: Unexpected exception _Map_base::at
Building engine failed

Failed to build CUDA engine on yolov5s.cfg
ERROR: Failed to create network using custom network creation function
ERROR: Failed to get cuda engine from custom library API
0:00:08.701327807 12280      0x16b5800 ERROR                nvinfer gstnvinfer.cpp:640:gst_nvinfer_logger:<primary_gie> NvDsInferContext[UID 1]: Error in NvDsInferContextImpl::buildModel() <nvdsinfer_context_impl.cpp:1934> [UID = 1]: build engine file failed
0:00:08.861269885 12280      0x16b5800 ERROR                nvinfer gstnvinfer.cpp:640:gst_nvinfer_logger:<primary_gie> NvDsInferContext[UID 1]: Error in NvDsInferContextImpl::generateBackendContext() <nvdsinfer_context_impl.cpp:2020> [UID = 1]: build backend context failed
0:00:08.861310525 12280      0x16b5800 ERROR                nvinfer gstnvinfer.cpp:640:gst_nvinfer_logger:<primary_gie> NvDsInferContext[UID 1]: Error in NvDsInferContextImpl::initialize() <nvdsinfer_context_impl.cpp:1257> [UID = 1]: generate backend failed, check config file settings
0:00:08.861354909 12280      0x16b5800 WARN                 nvinfer gstnvinfer.cpp:846:gst_nvinfer_start:<primary_gie> error: Failed to create NvDsInferContext instance
0:00:08.861366014 12280      0x16b5800 WARN                 nvinfer gstnvinfer.cpp:846:gst_nvinfer_start:<primary_gie> error: Config file path: /mnt/sd/workspace/customer/seeed/DeepStream-Yolo/config_infer_primary_yoloV5.txt, NvDsInfer Error: NVDSINFER_CONFIG_FAILED
** ERROR: <main:716>: Failed to set pipeline to PAUSED
Quitting
ERROR from primary_gie: Failed to create NvDsInferContext instance
Debug info: /dvs/git/dirty/git-master_linux/deepstream/sdk/src/gst-plugins/gst-nvinfer/gstnvinfer.cpp(846): gst_nvinfer_start (): /GstPipeline:pipeline/GstBin:primary_gie_bin/GstNvInfer:primary_gie:
Config file path: /mnt/sd/workspace/customer/seeed/DeepStream-Yolo/config_infer_primary_yoloV5.txt, NvDsInfer Error: NVDSINFER_CONFIG_FAILED
App run failed

Please help

[marcoslucianops]

Did you compile the nvdsinfer_custom_impl_Yolo lib with OpenCV support? What is the OpenCV version in your board?

[dharagg07]

Even I am Having the same error. I had installed the OpenCV version using the below command line even then it's showing the same error.

CUDA_VER=12.1 OPENCV=1 make -C nvdsinfer_custom_impl_Yolo

[IronmanVsThanos]

Even I am Having the same error. I had installed the OpenCV version using the below command line even then it's showing the same error.

CUDA_VER=12.1 OPENCV=1 make -C nvdsinfer_custom_impl_Yolo

have U sloved this problem?