No detection on Jetson Orin Nano

Accelerated Computing Intelligent Video Analytics DeepStream SDK
,
1

I have my own deepstream pipeline to detect vehicles. Before I was using Yolo v4 but then I decided to migrate to YoloX. After training, exporting and loading my onnx file into the system no detection at all. The pipeline whole pipeline was working because I was able to see the video streaming.
I tested on a Jetson Orin AGX and detection is working with the new network.

Jetson Orin Nano is using Jetpack 6.1 with Deepstream 7.1
Jeston Orin AGX is using Jetpack 6.0 with Deepstream 7.0

My config file is the following one (same for both system):

[property]
gpu-id=0
net-scale-factor=1.0
model-color-format=0
model-engine-file=../models/model_b1_gpu0_fp32.engine
onnx-file=../models/yolo-x/best_ckpt.onnx
labelfile-path=../models/yolo-x/labels.txt
batch-size=1
network-mode=0
num-detected-classes=1
interval=0
gie-unique-id=1
process-mode=1
network-type=0
cluster-mode=2
maintain-aspect-ratio=1
symmetric-padding=0
parse-bbox-func-name=NvDsInferParseYolo
custom-lib-path=../DeepStream-Yolo/nvdsinfer_custom_impl_Yolo/libnvdsinfer_custom_impl_Yolo.so
engine-create-func-name=NvDsInferYoloCudaEngineGet

[class-attrs-all]
nms-iou-threshold=0.45
pre-cluster-threshold=0.25
topk=300

Also FYI: I’m using Yolox Large model\

I used this script to port my model

using the following comand
python3 export_yolox.py -w best_ckpt.pth -c exps/default/ft_yolox_l.py --dynamic

4
  1. please make sure the model and nvinfer configuration file are the same on Nano and AGX. can deepstream-test1 run well on Orin Nano?
  2. where is “exps/default/ft_yolox_l.py”? could you try this guide to generate model? if nms-iou-threshold and pre-cluster-threshold are set to 0, does the issue persist? if so, please share the complete running log. Thanks!
5

There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks