visual_prompting
visual_prompting copied to clipboard
Official implementation and data release of the paper "Visual Prompting via Image Inpainting".
Dataset
Hi, would you please a downloaded dataset for training? Downloading from Arvix is very unstable and often forbidden.
I don't quite understand the model_without_ddp in the code, what does it do?
raise AttributeError("'{}' object has no attribute '{}'".format( AttributeError: 'Block' object has no attribute 'drop_path'
In the function ```calculate_metric(args, target, ours)```,there is a > assert colors.shape[0] == 2, colors # white and the expected color and I always got an error > File "/home/featurize/work/yjw/visual_prompting/evaluate/evaluate_reasoning.py", line...
Hello. I have a clarification question regarding input grids and support pairs. It looks like the model always works with a 224×224 input image, which is tokenized into 14×14 patches....