[TRT] jetson agx orion error - CaffeParser: Could not open file device GPU, failed to load networks/Googlenet/bvlc_googlenet.caffemodel

[sbggopal]

Hi team,

Am trying run inference of recognition program following the tutorial by buliding the repo, and i have installd all the utils and inference files , necessary to run the program. Also i have downloaded all the necessary model files with download_models.sh, and the model files are all in the ’ /usr/local/bin/networks/’ , but still while trying to run the example , am getting error as

[TRT] device GPU, loading /usr/local/bin/networks/Googlenet/bvlc_googlenet.caffemodel
[TRT] CaffeParser: Could not open file
[TRT] CaffeParser: Could not parse deploy file
[TRT] device GPU, failed to parse caffe network
[TRT] device GPU, failed to load networks/Googlenet/bvlc_googlenet.caffemodel
[TRT] failed to load networks/Googlenet/bvlc_googlenet.caffemodel
[TRT] imageNet – failed to initialize.
Traceback (most recent call last):
File “./recognition_inference.py”, line 44, in
net = imageNet(args.network)
Exception: jetson.inference – imageNet failed to load network

once i checked to the model files, all are there

Anyone have any idea , what might be the issue,

Thanks and regards
Bggopal

[sbggopal]

Note : - this only happens with , imagenet models, while trying to load bvlc_googlenet.caffemodel, testes with depthnet (ssd-mobilenet-v2 ) it works

is this a error, only doesn’t work for me

@dusty_nv

[SivaRamaKrishnaNV]

Dear @sbggopal ,
Could you test the model with trtexec to confirm if it is working ?

[sbggopal]

Hi @SivaRamaKrishnaNV

Thanks for the reply. trtexec is there in the /usr/share/tensorrt/bin . but not detected on terminal, when i run
‘trtexec --help’
Am checking on that. will check let you know

Thanks and regards
Bggopal