How to run lprnet with DLA?

Accelerated Computing Intelligent Video Analytics TAO Toolkit
#1

Please provide the following information when requesting support.

• Hardware:Xavier NX
• Network Type: LPRnet
• TLT Version: format_version: 1.0 toolkit_version: 3.21.08 published_date: 08/17/2021

I can successfully run the LPRnet sample code, and I want to use DLA in my Jetson Xavier XN.

How to run lprnet with using DLA?

#3

You can enable dla in deepstream config file. For example, in deepstream_lpr_app/lpd_us_config.txt at b5447b7174f772b4f855f576cb82d33378768fb6 · NVIDIA-AI-IOT/deepstream_lpr_app · GitHub

enable-dla=1
use-dla-core=0

#4

However, the LPRnet sample code which I use is python-based. Could you show where is the option to place the configure file.

https://forums.developer.nvidia.com/t/not-getting-correct-output-while-running-inference-using-tensorrt-on-lprnet-fp16-model/184396

#5

One relevant topic in Tensorrt forum, Accessing Jetson's DLA from python

#7

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