Where are DNN model files for driveworks 3.5 DNN sample application?

Please provide the following info (check/uncheck the boxes after clicking “+ Create Topic”):
Software Version
DRIVE OS Linux 5.2.0
DRIVE OS Linux 5.2.0 and DriveWorks 3.5
NVIDIA DRIVE™ Software 10.0 (Linux)
NVIDIA DRIVE™ Software 9.0 (Linux)
other DRIVE OS version

Target Operating System

Hardware Platform
NVIDIA DRIVE™ AGX Xavier DevKit (E3550)
NVIDIA DRIVE™ AGX Pegasus DevKit (E3550)

SDK Manager Version

Host Machine Version
native Ubuntu 18.04

I try to run Driveworks 3.5 sample application “sample_dnn_tensor” on my custom board with Xavier. it says " INVALID_CONFIG: Using an engine plan file across different models of devices is not recommended and is likely to affect performance or even cause errors. "

I had a look similar forums and they all say I need to rebuild engine file with “tensorRT_optimization” tool. By the way, nobody says where are uff or caffe or onnx files to run “tensorRT_optimization”. I guess there should be corresponding model file for “sample_dnn_tensor” and other dnn sample application for driveworks 3.5.

I want to use exactly same model file which are used with “sample_dnn_tensor” or other driveworks 3.5 dnn sample application as a “Golden sample”.

It does not make sense that there is no model file and Nvidia is keeping saying “You need to rebuild engine plan file on your target.”.

Does anybody know?

Dear @youngsouck.eun,
May I know which platform are you using and confirm if you flashed the target using sdkmanager? Also, could you cross compile CUDA devicequery sample, run on the target and share output log

Hello SivaRamaKrishnaNV,

I use custom board based on e3550. I did not flash with sdkmanager and used command line to run bootburn.

After flashing and booting, I installed driveworks3.5 on my target board and could run 2 sample driveworks application “sample_camera” and “sample_denseopticalflow”.

Here comes result of “deviceQuery”.

nvidia@tegra-ubuntu:~$ ./deviceQuery
./deviceQuery Starting…

CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 1 CUDA Capable device(s)

Device 0: “Xavier”
CUDA Driver Version / Runtime Version 10.2 / 10.2
CUDA Capability Major/Minor version number: 7.2
Total amount of global memory: 31537 MBytes (33069203456 bytes)
( 8) Multiprocessors, ( 64) CUDA Cores/MP: 512 CUDA Cores
GPU Max Clock rate: 1109 MHz (1.11 GHz)
Memory Clock rate: 1109 Mhz
Memory Bus Width: 256-bit
L2 Cache Size: 524288 bytes
Maximum Texture Dimension Size (x,y,z) 1D=(131072), 2D=(131072, 65536), 3D=(16384, 16384, 16384)
Maximum Layered 1D Texture Size, (num) layers 1D=(32768), 2048 layers
Maximum Layered 2D Texture Size, (num) layers 2D=(32768, 32768), 2048 layers
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 49152 bytes
Total number of registers available per block: 65536
Warp size: 32
Maximum number of threads per multiprocessor: 2048
Maximum number of threads per block: 1024
Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
Max dimension size of a grid size (x,y,z): (2147483647, 65535, 65535)
Maximum memory pitch: 2147483647 bytes
Texture alignment: 512 bytes
Concurrent copy and kernel execution: Yes with 1 copy engine(s)
Run time limit on kernels: No
Integrated GPU sharing Host Memory: Yes
Support host page-locked memory mapping: Yes
Alignment requirement for Surfaces: Yes
Device has ECC support: Disabled
Device supports Unified Addressing (UVA): Yes
Device supports Compute Preemption: Yes
Supports Cooperative Kernel Launch: Yes
Supports MultiDevice Co-op Kernel Launch: Yes
Device PCI Domain ID / Bus ID / location ID: 0 / 0 / 0
Compute Mode:
< Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 10.2, CUDA Runtime Version = 10.2, NumDevs = 1
Result = PASS

Dear @youngsouck.eun ,
The above warning coomes when we try to load a different engine not generated for specific GPU.
TRT 6.3 considers the warning " INVALID_CONFIG: Using an engine plan file across different models of devices is not recommended and is likely to affect performance or even cause errors" as an error and blocks deserialization. It is possible that your board GPU parameters are different than DRIVE Xavier Developer platform. These models are prepared for DRIVE Xavier Developer platform and hence you notice above warning.

Driveworks package contains only TRT model binary to demonstrate sample applications.

Do you know where is model file to generate model binary for driveworks 3.5 DNN examples?

Dear @youngsouck.eun,
The model file is not sharable on public forum. Could you check with your NVIDIA representative on this? I will also check internally on how this can be shared.

Dear @youngsouck.eun ,
It seems the the caffemodel file for sample_dnn_tensor is already shared in DW package at data/samples/detector. Could you use that and check?

Dear @SivaRamaKrishnaNV,

I checked the location and tried to generate plan file and run application.

It works. Thank you very much. You should be much better than representative who is located in San Jose.