Get wrong input and output network infomation

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

Hello,My network has three inputs.their shpaes are 1x3x640x960, 1x3x3,and 1x2.Output shape is 50x14.
Screenshot from 2022-12-27 10-14-21

However the “Implicit Engine Info” shows that the inputs dimension are 3x640x960,3x3,2.The output dimension is 14.

Screenshot from 2022-12-27 10-30-41
Screenshot from 2022-12-27 10-30-41726×159 26.2 KB

It turns out that the first dimension disappear.How can I solve this problem ,thanks

2

Moved to TensorRT forum for better support.

5

Hi,

Could you please share with us the TensorRT version you’re using and ONNX model to try from our end for better debugging.

Thank you.