UffParser: Unsupported number of graph 0

It is not expected to run against mp4 file. An h264 file is expected.

Can you share your latest “ll /usr/lib/x86_64-linux-gnu/libnvinfer_plugin.so*” ?

Even an RTSP stream is also making no difference, and what about a jpg image??

A jpg image can be used.

That is what I am trying to explain that even a .jpg image when given as an argument throws error

Other users can run. So, there should be something wrong on your setup environment.

Can you share your latest “ll /usr/lib/x86_64-linux-gnu/libnvinfer_plugin.so*” ?

I have shared this…

-rwxr-xr-x 1 root root 3711136 Mar 17 13:39 /usr/lib/x86_64-linux-gnu/libnvinfer_plugin.so*
lrwxrwxrwx 1 root root 26 Mar 17 13:55 /usr/lib/x86_64-linux-gnu/libnvinfer_plugin.so.7 → libnvinfer_plugin.so.7.2.2*
-rwxr-xr-x 1 root root 3711136 Mar 17 12:13 /usr/lib/x86_64-linux-gnu/libnvinfer_plugin.so.7.0.0*
-rwxr-xr-x 1 root root 4829728 Mar 17 13:55 /usr/lib/x86_64-linux-gnu/libnvinfer_plugin.so.7.2.2*

It looks not expected.
You can see the expected similar to Failling in building sample from TLT-DEEPSTREAM - #15 by Morganh .
But please note that it is for Jetpack devices.

For you device, which dgpu are you using?
Do you have the original status for “ll /usr/lib/x86_64-linux-gnu/libnvinfer_plugin.so*” ?

I am not running on any Jetpack devices.

I know you are running with dpgu device.
Please note that when you run below step mentioned in

sudo mv /usr/lib/x86_64-linux-gnu/libnvinfer_plugin.so.8.p.q ${HOME}/libnvinfer_plugin.so.8.p.q.bak

The original one(I think it is /usr/lib/x86_64-linux-gnu/libnvinfer_plugin.so.7.0.0*) will be moved to your home folder. But I can still see the /usr/lib/x86_64-linux-gnu/libnvinfer_plugin.so.7.0.0 in the /usr/lib/x86_64-linux-gnu/ folder.

Let me check that

I am unable to find a link to download DeepStream-5.1…
I am being redirected to 6.0.1

OK, let me clarify since the user guide may confuse you. We will improve it.

For 6.0.1 + yolov4-tiny, you can run either .etlt model or tensorrt engine with Deepstream.

For 5.1 + yolov4-tiny, you can only run tensorrt engine with deepstream.

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