We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
e753646
Pre-trained attention model, corresponding mapping models, and a ResNet-34 (multi-label classification for Danbooru2020) for reference encoder.
