How to find the input name of EmotionNet?

i tried to convert EmotionNet(EmotionNet | NVIDIA NGC) toTensorRT from model.etlt

./tlt_converter model.etlt -k nvidia_tlt -p input,1x1x136x1,1x1x136x1,1x1x136x1

the input name is wrong, how can I find the correct input name?

Please refer to below.

tao-converter model.etlt
-k nvidia_tlt
-t fp32
-p input_landmarks:0,1x1x136x1,1x1x136x1,2x1x136x1
-e model.engine

the question is, how did you find that?

Please refer to TAO Computer Vision Inference Pipeline | NVIDIA NGC

and then download and view below file.
wget --content-disposition ‘https://api.ngc.nvidia.com/v2/resources/nvidia/tao/tao_cv_inference_pipeline_quick_start/versions/v0.3-ga/files/scripts/tao_cv_compile.sh