AK51
June 8, 2022, 6:17am
1
Hi,
I want to train the custom object detection in AGX and move the onnx to nano, but the nano stuck at
“[TRT] Could not register plugin creator - ::FlattenConcat_TRT version 1”
Other than the
labels.txt
ssd-mobilenet.onnx
in the models folder, what do I need to move to the nano?
Thx
AK51
June 8, 2022, 6:26am
2
Hi,
If I add this file to the models folder
ssd-mobilenet.onnx.1.1.8001.GPU.FP16.engine
It said hashRead failed.CRC-32
Thx
Hi,
Do you use the same source and TensorRT version on Nano?
The output indicates some implementation is missing.
Thanks.
AK51
June 8, 2022, 7:36am
5
Hi,
Yes, I did rename the existing jetson-inference folder to jetson-inference-old in both AGX and nano,
and git clone the inference and build them
Thx
Hi,
Does the inference start after the message?
Based on the discussion on GitHub, the error occurs in the first build.
opened 08:55PM - 05 Dec 20 UTC
closed 09:09PM - 07 Dec 20 UTC
I am using the video tutorial on Youtube to develop my first real time object de… tection. When I try to run the script, the following error show up `[TRT] Could not register plugin creator - ::FlattenConcat_TRT version 1
`
After the error, the camera never turns on and the terminal keeps on spitting values like this:
After concat removal: 66 layers
[TRT] Graph construction and optimization completed in 0.045886 seconds.
[TRT] Constructing optimization profile number 0 [1/1].
[TRT] --------------- Timing Runner: <reformat> (Reformat)
[TRT] Tactic: 1002 time 0.747864
[TRT] Tactic: 0 time 1.23659
[TRT] Fastest Tactic: 1002 Time: 0.747864
[TRT] --------------- Timing Runner: <reformat> (Reformat)
[TRT] Tactic: 1002 time 20.0891
[TRT] Tactic: 0 time 1.31471
[TRT] Fastest Tactic: 0 Time: 1.31471
Has anyone come up with the same issue and knows how to fix it? I'm running the script on the Jetson Nano 4GB Development kit with Jetpack 4.4 installed and using the Raspberry Pi Camera V2.
This is the actual script:
```import jetson.inference
import jetson.utils
net = jetson.inference.detectNet("coco-dog", threshold=0.5)
camera = jetson.utils.videoSource("csi://0") # '/dev/video0' for V4L2
display = jetson.utils.videoOutput("display://0") # 'my_video.mp4' for file
while display.IsStreaming():
img = camera.Capture()
detections = net.Detect(img)
display.Render(img)
display.SetStatus("Object Detection | Network {:.0f} FPS".format(net.GetNetworkFPS()))```
Thanks.
AK51
June 16, 2022, 10:58am
7
Hi,
I used the pth file from AGX and did the onnx conversion in nano and it works now.
Thx
system
Closed
July 6, 2022, 7:46am
9
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.