segmenteverygrain
segmenteverygrain copied to clipboard
Train without creating patches from large images?
Hello, thanks for the repo! Any chance I can train/finetune the model without resorting to creating patches?