Something is wrong with the paths, may it be due to using a SDD with Jetson Xavier NX?

[Shemesu]

Something is wrong with the paths, may it be due to using a SDD with Jetson Xavier NX?

I run the docker and it works, but to run the imagenet I have to add the path of the files:

./imagenet.py ../../data/images/jellyfish.jpg ../../data/images/test/output_0.jpg

But then I thried the posenet and, even with the full path, or even adding the network it doesn't work:

root@shemesu-desktop:/jetson-inference/python/examples# ./posenet.py "../../data/images/humans_*.jpg" ../../data/images/test/posehumans%i.jpg

poseNet -- loading pose estimation model from: -- model networks/Pose-ResNet18-Body/pose_resnet18_body.onnx -- topology networks/Pose-ResNet18-Body/human_pose.json -- colors networks/Pose-ResNet18-Body/colors.txt -- input_blob 'input' -- output_cmap 'cmap' -- output_paf 'paf' -- threshold 0.150000 -- batch_size 1

[TRT] poseNet -- failed to load topology json from 'networks/Pose-ResNet18-Body/human_pose.json' Segmentation fault (core dumped)

OR

root@shemesu-desktop:/jetson-inference/python/examples# ./posenet.py --network="resnet18-body" "../../data/images/humans_*.jpg" ../../data/images/test/posehumans%i.jpg

poseNet -- loading pose estimation model from: -- model networks/Pose-ResNet18-Body/pose_resnet18_body.onnx -- topology networks/Pose-ResNet18-Body/human_pose.json -- colors networks/Pose-ResNet18-Body/colors.txt -- input_blob 'input' -- output_cmap 'cmap' -- output_paf 'paf' -- threshold 0.150000 -- batch_size 1

[TRT] poseNet -- failed to load topology json from 'networks/Pose-ResNet18-Body/human_pose.json' Segmentation fault (core dumped)

I'm running it in L4T 32.5.1 in a Jetson Xavier NX with SDD drive.

Thanks,

[dusty-nv]

Hmm... is the SSD formatted as NTFS/FAT as opposed to ext3/ext4? If so, that may explain why the symbolic links aren't working for you.