ResizeNearest layer get error on deserialize

Autonomous Vehicles DRIVE AGX Xavier DRIVE AGX Xavier General
, ,
1

Hardware Platform: x86 PC with 2080Ti GPU
Software Version: DRIVE Software 10
Host Machine Version: Ubuntu 18.04
SDK Manager Version: 1.1.0-6343

pytorch version: 1.0.0

Hello,

I am getting a error when deserialize a model that have ResizeNearest layer.
We can generate the “optimized.bin” file from tensorRT_optimization software. But when we loading the “optimized.bin” file in our software, it showing Cannot deserialize plugin ResizeNearest:

image
image780×186 9.82 KB

I uploaded the onnx file and the python script for generating the model in the model.zip file
model.zip (945 Bytes)

The same dnn layer works in DriveOS 9.
Could you help us solve this problem?

Thanks

2

Dear @qia7r6b,
Could you check Using bin-file after TensorRT optimization

4

Dear @qia7r6b,
The same issue is discussed in the mentioned thread. Marking this thread as duplicate.