What should I prepare, if I want to train network on my own images? I'm not sure I understand everything right. I have images 300x300, and I can create binary masks for these images, but I suppose it is not all, I saw some distance and size files when I run python main.py prepare-masks. Can somebody explain algorythm of preparing own data for training?
What should I prepare, if I want to train network on my own images? I'm not sure I understand everything right. I have images 300x300, and I can create binary masks for these images, but I suppose it is not all, I saw some distance and size files when I run
python main.py prepare-masks
. Can somebody explain algorythm of preparing own data for training?