Question about quantifying in xavier

Autonomous Vehicles DRIVE AGX Xavier DRIVE AGX Xavier General
1

Please provide the following info (check/uncheck the boxes after creating this topic):
Software Version
DRIVE OS Linux 5.2.6
DRIVE OS Linux 5.2.0
[y] 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
other

Target Operating System
[y] Linux
QNX
other

Hardware Platform
[y] NVIDIA DRIVE™ AGX Xavier DevKit (E3550)
NVIDIA DRIVE™ AGX Pegasus DevKit (E3550)
other

SDK Manager Version
[y] 1.6.0.8170
other

Host Machine Version
[y] native Ubuntu 18.04
other

Usually, we use 512 images to quantify model by tensorrt.
When I quantify in Xavier in GPU or DLA, I failed because of no enough store.
My each image size is 192012803*4, and I have 512 images.
By my watch, when test read image file, there are about 14000 MB left, but then they are used quikly and test goes to be failed.

2

Hi @wang_chen2,

Can you try with SSD USB drive?

3

Hi, @VickNV,

How?
The calb data is in another server mounted to xavier.
SSD USB drive ?

4

Can’t you mount a larger storage. Or I don’t get your point.

5

My 512 images data is 14G and I save it in A server then mount it to Xavier.
Then in Xavier, do quantify, the RAM showed in tegrastatus will run out.

6

Per Xaiver has only 32GB LPDDR4x system ram.

7

Hi, I have solved this question by modify code.
Thank u very much.