PyTorch training engine for urban perception modeling. Uses ViT multi-head architecture to learn human preferences from pairwise comparisons, with TrueSkill-regularized ranking loss and Optuna HPO.
machine-learning computer-vision deep-learning urban-planning trueskill pytorch preference-learning pairwise-ranking optuna vision-transformer dinov2 urban-perception
-
Updated
Apr 28, 2026 - Python