Failed to create NvDsInferContext instance

  • yolov5-5.0
  • jetpack-4.4
  • deepstream-5.0
  • Tensorrt-7.1
  • Cuda-10.2

Perform the last step:LD_PRELOAD=./libmyplugins.so deepstream-app -c deepstream_app_config_yoloV5.txt

I received the following error:

  1. ERROR from primary_gie: Failed to create NvDsInferContext instance

  2. Debug info: /dvs/git/dirty/git-master_linux/deepstream/sdk/src/gst-plugins/gst-nvinfer/gstnvinfer.cpp(809): gst_nvinfer_start (): /GstPipeline:pipeline/GstBin:primary_gie_bin/GstNvInfer:primary_gie:

  3. Config file path: /home/yqy/workspace/Yolov5-mask-detect-in-Deepstream-5.0-tensorrt7/Deepstream 5.0/config_infer_primary_yoloV5.txt, NvDsInfer Error: NVDSINFER_CONFIG_FAILED

  4. App run failed

Sorry for the late response, is this still an issue to support? Thanks

Hi,I successfully deployed my model but the frame rate was very low, only 7 frames, I used logitech’s c920 model, the parameters are as follows:[source0]
enable=1
#Type - 1=CameraV4L2 2=URI 3=MultiURI
type=1
camera-width=1280
camera-height=720
camera-fps-n=30
camera-v4l2-dev-node=0
#uri=file://…/…/samples/streams/sample_1080p_h264.mp4
num-sources=1
gpu-id=0

(0): memtype_device - Memory type Device

(1): memtype_pinned - Memory type Host Pinned

(2): memtype_unified - Memory type Unified

cudadec-memtype=0

What precision or format should I use to improve the frame rate, thanks

what’s the HW platform?
What inference precision did you choose?

you can use command like below to check the model perf
# /usr/src/tensorrt/bin/trtexec --loadEngine=TRT_Engine_Generated_by_DeepStream --plugins=./libmyplugins.so

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.