Hi
I have used the two images in my training.
I have used the resnet50-v2-7, which I successfully converted to a .tltb file to use in byom and I am using the byom notebook. I did this to be 100% sure I was starting a new Jupyter Notebook with no old code residing:
I used your mobilenet_v2 training_spec.cfg (renamed “morganh_spec.cfg”):
tao classification train still throws the PIL.UnidentifiedImageError
You get no error
I have repeated the process using the exact cat image from your steps:
$ wget https://global.discourse-cdn.com/nvidia/original/3X/7/b/7bb45815e373723b77f592895b4fab7ee8e72747.jpeg
and get the same result (error).
I don’t know what else I can try.
Please advise
