ProbeX/Model-J__SupViT__model_idx_0008

Model-J: SupViT Model (model_idx_0008)

This model is part of the Model-J dataset, introduced in:

Learning on Model Weights using Tree Experts (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen

🌐 Project | 📃 Paper | 💻 GitHub | 🤗 Dataset

Model Details

Attribute	Value
Subset	SupViT
Split	train
Base Model	google/vit-base-patch16-224
Dataset	CIFAR100 (50 classes)

Training Hyperparameters

Parameter	Value
Learning Rate	9e-05
LR Scheduler	constant_with_warmup
Epochs	6
Max Train Steps	1998
Batch Size	64
Weight Decay	0.007
Seed	8
Random Crop	False
Random Flip	False

Performance

Metric	Value
Train Accuracy	0.9940
Val Accuracy	0.9373
Test Accuracy	0.9292

Training Categories

The model was fine-tuned on the following 50 CIFAR100 classes:

chimpanzee, boy, trout, girl, porcupine, lion, pear, man, television, leopard, bee, bear, rocket, woman, crocodile, keyboard, seal, table, turtle, lamp, cloud, ray, hamster, pickup_truck, plain, whale, maple_tree, bus, telephone, shark, sweet_pepper, aquarium_fish, bed, beetle, rose, rabbit, apple, dinosaur, house, bottle, tank, snail, crab, forest, lawn_mower, worm, chair, lizard, fox, tiger