TensorRT: Mobilenet v2 onnx model not able to convet to TRT with TensorRT 8

@pranavm-nvidia based on your comments on this issue: https://github.com/NVIDIA/TensorRT/issues/1379

I tried to convert the mobilenet v2 model to trt as earlier it had an issue of unsupported NMS layers which is now supported in tensorRT 8, thus we don’t need to add any custom layers.

I followed the same process as suggested in the issue 1379

1. Convert the Tflite model to ONNX

python -m tf2onnx.convert --opset 11 --tflite mv2_float.tflite --output mv2_90cls_11_tflite_model.onnx

1. Convert the onnx model to trt trtexec --onnx=4_Oct/mv2_90cls_11_tflite_model.onnx --saveEngine=4_Oct/mv_90cls_Engine_new.trt --explicitBatch --verbose > logs.txt

We get the below error

The tflite and onnx model are shared in the gdrive link

logs.txt