Hello,
I’m trying to train from my own dataset. I have planty of pictures in jpg/jpeg format which I want to use to train my own model - how to do it?
Hi @nerk, are you trying to train image classifier or object detector?
If you are doing classification, you can simply organize your data like shown here, where each class is in it’s own subfolder:
https://github.com/dusty-nv/jetson-inference/blob/master/docs/pytorch-collect.md
‣ train/
• class-A/
• class-B/
• ...
‣ val/
• class-A/
• class-B/
• ...
‣ test/
• class-A/
• class-B/
• ...
If you are doing object detection, I recommend using CVAT tool, and then after you are done annotating, export it in Pascal VOC format. Then add a labels.txt file to your extracted dataset.