Setup error while running demo app

replytosubodh · November 30, 2020, 11:15am

Please provide complete information as applicable to your setup.

• Hardware Platform (dGPU) T4
• DeepStream Version 5.01
• TensorRT Version 7
• NVIDIA GPU Driver Version (valid for GPU only) 450
• Issue Type( questions, new requirements, bugs) question
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)

pulled docker and tried to configure manually in both cases facing pipeline build failed issue.
deepstream is installed correctly and deepstream --help runs successfully

Fiona.Chen · December 1, 2020, 1:33am

Which docker are you using? What is your command to run the docker image? Do you know the CUDA version in your host?

Have you checked the folder of /usr/lib/x86_64-linux-gnu/gstreamer-1.0/deepstream in your docker? Does it look like

What will happen if you run the following command?
gst-inspect-1.0 | grep nvinfer

replytosubodh · December 1, 2020, 9:16am

I am using nvcr.io/nvidia/deepstream docker.

Have you checked the folder of /usr/lib/x86_64-linux-gnu/gstreamer-1.0/deepstream in your docker? Does it look like
Yes it looks like screenshot you shared.

cuda version is V9.1.85

on running gst-inspect-1.0 | grep nvinfer i attached output in screenshot.

Fiona.Chen · December 1, 2020, 11:26am

DeepStream needs CUDA10.2 or higher version.

replytosubodh · December 1, 2020, 12:14pm

Installed 10.2 but still same problem.

Fiona.Chen · December 2, 2020, 1:06am

The link is the instruction of DeepStream containers. Please following it and try.
Docker Containers — DeepStream DeepStream Version: 5.0 documentation (nvidia.com)

replytosubodh · December 3, 2020, 7:51am

Getting this error

replytosubodh · December 3, 2020, 7:52am

can you sent latest installation documentation

Fiona.Chen · December 3, 2020, 8:04am

Have you run “sudo apt-get update”? The documentation is already the latest one.

replytosubodh · December 3, 2020, 9:50am

Followed the guide and now getting this error -

Fiona.Chen · December 3, 2020, 10:41am

Can you put your command line? Which sample app are you using?

replytosubodh · December 3, 2020, 1:17pm

I am running deepstream-app -c and first .txt configuration file from config folder.

replytosubodh · December 3, 2020, 2:41pm

Fiona.Chen · December 8, 2020, 3:07am

“first”???Is it “config_infer_primary.txt”? “config_infer_primary.txt” is not deepstream-app config file but the nvinfer config file. “source4_1080p_dec_infer-resnet_tracker_sgie_tiled_display_int8.txt” is deepstream-app config file, you can try it.

Please refer to https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_sample_configs_streams.html

It is highly recommanded to study the deepstream document before you run any sample. https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_C_Sample_Apps.html

replytosubodh · December 13, 2020, 10:32am

I bought Jetson Nano and I could run basic mobile ssd v2 on jetson nano.
I want to run dashcamnet which is pretrained model by nvidia.
Please guide on it.

Fiona.Chen · December 13, 2020, 1:52pm

You can refer to deepstream_app_source1_dashcamnet_vehiclemakenet_vehicletypenet.txt in /opt/nvidia/deepstream/deepstream-5.0/samples/configs/tlt_pretrained_models/

This sample use dashcamnet as PGIE, use vehiclemakenet and vehicletypenet as SGIEs.

replytosubodh · December 13, 2020, 3:27pm

Can you please guide more .
I want to run dashcam model urgently .
Please provide detailed steps and tell if there is something that need to be downloaded for example tlt and etlt model and their directory.

replytosubodh · December 13, 2020, 5:18pm

I want to use dashcam model and want to add some code after detections.Please guide for this.