UffParser: Validator error: FirstDimTile_2: Unsupported operation _BatchTilePlugin_TRT

AI & Data Science Deep Learning (Training & Inference) TensorRT
#1

Description

I trained the yolov3 model on TLT3.0 and encountered this error when deploying the model to deepstream. How can I solve it? Thanks

image
image1717×165 13 KB

Environment

train
TensorRT Version: 7.2.1.6
GPU Type: 3080
Nvidia Driver Version: 460.39
CUDA Version: 10.2
CUDNN Version: 8.1.0
Operating System + Version: Ubuntu18.04
Python Version (if applicable): 3.6.9
TLT：3.0
deploy:
TensorRT Version: 7.2.1.6
GPU Type: V100
Nvidia Driver Version: 450.80.02
CUDA Version: 10.2
CUDNN Version: 8.1.1
Operating System + Version: Ubuntu18.04
Python Version (if applicable): 3.6.9
DeepStream：5.0.0

[TRT]: UffParser: Unsupported operation _BatchTilePlugin_TRT
#2

Hi , UFF and Caffe Parser have been deprecated from TensorRT 7 onwards, hence request you to try ONNX parser.
Please check the below link for the same.

Thanks!

#3

Thank you.
I am using the files under the path of /deepstream-5.0/sources/objectDetector_Yolo, how can I modify it to use ONNX parser?

#4

Hi @1210586191,

This looks like DeepStream related issue, we request you to post your concern here for better help.

Thank you.

#5

Thanks for your reply!

I encountered an error when running deepstream-app.

image
image909×22 1.99 KB
Is it a version problem?
Is deepstream5.0 compatible with TensorRT7.2.1.6?

#6

Hi @1210586191,

Yes there is a compatibility. This doc may help you.
https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_Quickstart.html#platform-and-os-compatibility

For further assistance on error, please post your query on DeepStream forum link shared above.

Thank you.