The problem of the key used to load the model is incorrect

Accelerated Computing Intelligent Video Analytics TAO Toolkit
1

Hello. When I tried to export the cal_cache_file of resnet50_peoplenet_transformer.tlt, I got the error message displayed that the key used to load the model is incorrect.

However. The model key I set is nvidia_tao from official website.

How could I do to deal with the exporting problem ?

The contents below are my command to export cache file|

sudo docker run -it --rm -v /home/ubuntu/tao_test_2023/tensorflow_train/model_zoo/detectnet_v2/peoplenet_transformer/:/workspace/tao-experiments 
                         nvcr.io/nvidia/tao/tao-toolkit:4.0.0-tf1.15.5 detectnet_v2 export 
                         -m /workspace/tao-experiments/resnet50_peoplenet_transformer.tlt 
                         -o /workspace/tao-experiments/result2 
                         --cal_cache_file /workspace/tao-experiments/result2 
                         -k nvidia_tao

The picture below is the error message

image
image1408×741 68.6 KB

2

Can you share the website?

3

PeopleNet Transformer | NVIDIA NGC

4

Can you try nvidia_tlt or tlt_encode ?

5

It works by changing the key to tlt_encode. Thank you very much !

6

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.