Tried Deepstream sample deepstream_test_1.py but failed

Autonomous Machines Jetson & Embedded Systems Jetson Nano
1

Dear Sir,
I try to execute deepstream using the following instruction :
$ python3 deepstream_test_1.py HK002.mp4

But got the following error message :

Using winsys: x11
Opening in BLOCKING MODE
Creating LL OSD context new
0:00:04.150498450 12857 0x3349ad90 INFO nvinfer gstnvinfer.cpp:519:gst_nvinfer_logger: NvDsInferContext[UID 1]:initialize(): Trying to create engine from model files
0:00:04.150791999 12857 0x3349ad90 WARN nvinfer gstnvinfer.cpp:515:gst_nvinfer_logger: NvDsInferContext[UID 1]:generateTRTModel(): INT8 not supported by platform. Trying FP16 mode.
0:00:04.150848199 12857 0x3349ad90 ERROR nvinfer gstnvinfer.cpp:511:gst_nvinfer_logger: NvDsInferContext[UID 1]:generateTRTModel(): Cannot access caffemodel file ‘/home/astro/deepstream/t1/…/…/…/…/samples/models/Primary_Detector/resnet10.caffemodel’
0:00:04.150898044 12857 0x3349ad90 ERROR nvinfer gstnvinfer.cpp:511:gst_nvinfer_logger: NvDsInferContext[UID 1]:initialize(): Failed to create engine from model files
0:00:04.150947263 12857 0x3349ad90 WARN nvinfer gstnvinfer.cpp:692:gst_nvinfer_start: error: Failed to create NvDsInferContext instance
0:00:04.150970702 12857 0x3349ad90 WARN nvinfer gstnvinfer.cpp:692:gst_nvinfer_start: error: Config file path: dstest1_pgie_config.txt, NvDsInfer Error: NVDSINFER_CONFIG_FAILED
Error: gst-resource-error-quark: Failed to create NvDsInferContext instance (1): /dvs/git/dirty/git-master_linux/deepstream/sdk/src/gst-plugins/gst-nvinfer/gstnvinfer.cpp(692): gst_nvinfer_start (): /GstPipeline:pipeline0/GstNvInfer:primary-inference:
Config file path: dstest1_pgie_config.txt, NvDsInfer Error: NVDSINFER_CONFIG_FAILED

1.Is there anythin I haven’t installed?

2.Can someone provide an instruction that can execute the example,let me make sure that the execution environment is correct.

Thank you.

3

Hi
you need h264 elementary source.
To run the test app:
$ python3 deepstream_test_1.py <h264_elementary_stream>

1 Like
4

How do I adjust if I want to use h265?
Thank you.

deepstream_test_1.py

5

test3 use uridecodebin which support H265 codec, you can try with test3 sample.

1 Like
6
  1. I tried to execute test3 sample, but got the following error message:
    ==========================================================
    $ python3 deepstream_test_3.py rtsp://admin:aa12345678@192.168.2.4:554/h265/ch1/main/av_stream
    Creating Pipeline

Creating streamux

Creating source_bin 0

Creating source bin
source-bin-00
Creating Pgie

Unable to create pgie
Creating tiler

Creating nvvidconv

Creating nvosd

Creating transform

Creating EGLSink

Traceback (most recent call last):
File “deepstream_test_3.py”, line 346, in
sys.exit(main(sys.argv))
File “deepstream_test_3.py”, line 285, in main
pgie.set_property(‘config-file-path’, “dstest3_pgie_config.txt”)
AttributeError: ‘NoneType’ object has no attribute ‘set_property’

==========================================================

  1. I still get an error when i execute the following command:
    :~/deepstream_sdk_v4.0.2_jetson/samples/configs/deepstream-app$ deepstream-app -c source4_1080p_dec_infer-resnet_tracker_sgie_tiled_display_int8.txt
    deepstream-app: error while loading shared libraries: libnvinfer.so.6: cannot open shared object file: No such file or directory

==========================================================

  1. I tried to follow the content below, but still can’t solve it, how can i eliminate these problems?

Thank you.

tensorflow.python.framework.errors_impl.NotFoundError: libnvinfer.so.5: cannot open shared object file: No such file or directory

7

Traceback (most recent call last):

File “deepstream_test_3.py”, line 346, in
sys.exit(main(sys.argv))
File “deepstream_test_3.py”, line 285, in main
pgie.set_property(‘config-file-path’, “dstest3_pgie_config.txt”)
AttributeError: ‘NoneType’ object has no attribute ‘set_property’

I still get an error when i execute the following command:

:~/deepstream_sdk_v4.0.2_jetson/samples/configs/deepstream-app$ deepstream-app -c source4_1080p_dec_infer-resnet_tracker_sgie_tiled_display_int8.txt
deepstream-app: error while loading shared libraries: libnvinfer.so.6: cannot open shared object file: No such file or directory

→ make sure you got right version TensorRT installed

8

I thougth TensorRT was the default installation project, obviously not.

I use JetPack4.3, the current version is as follows:

(1)Jetson Nano Developer Edition

(2)$ cat /etc/issue
Ubuntu 18.04.4 LTS \n \l

(3)$ cat /usr/local/cuda/version.txt
CUDA Version 10.2.89

===============================================================

I thought i should download “nv-tensorrt-repo-ubuntu1804-cuda10.2-trt6.0.1.8-ga-20191108_1-1_amd64.deb”
But it cannot be installed.

Which version should be download?

Thank you.

9

this package is for x86 platform, if your device is jetson, sdkmanager will install whole for you if you choose all, os and software, I remember for software default if you choose, it will install all, i see you have cuda installed, not sure why can’t you find TensorRT, any output for this,
dpkg -l|grep infer

1 Like
10

Dear Sir,

$ dpkg -l|grep infer
ii libnvinfer-bin 7.1.0-1+cuda10.2 arm64 TensorRT binaries
ii libnvinfer-dev 7.1.0-1+cuda10.2 arm64 TensorRT development libraries and headers
ii libnvinfer-doc 7.1.0-1+cuda10.2 all TensorRT documentation
ii libnvinfer-plugin-dev 7.1.0-1+cuda10.2 arm64 TensorRT plugin libraries
ii libnvinfer-plugin7 7.1.0-1+cuda10.2 arm64 TensorRT plugin libraries
ii libnvinfer-samples 7.1.0-1+cuda10.2 all TensorRT samples
ii libnvinfer7 7.1.0-1+cuda10.2 arm64 TensorRT runtime libraries
ii python-libnvinfer 7.1.0-1+cuda10.2 arm64 Python bindings for TensorRT
ii python-libnvinfer-dev 7.1.0-1+cuda10.2 arm64 Python development package for TensorRT
ii python3-libnvinfer 7.1.0-1+cuda10.2 arm64 Python 3 bindings for TensorRT
ii python3-libnvinfer-dev 7.1.0-1+cuda10.2 arm64 Python 3 development package for TensorRT

=================================

It seems that TensorRT 7.1 is installed.

Te error message is about libnvinfer.so.6, does it mean that the current version i too new, so it’s not supported?

11

Jetpack 4.3 with TensorRT version 6.0, version 4.4 with TensorRT 7.1, you used Jetpack 4.4 i guess, but DS 5.0 not released, it will be soon, you can wait for new DS release or you can reflash Jetpack 4.3

1 Like