ProbeX/Model-J__SupViT__model_idx_0007

Model-J: SupViT Model (model_idx_0007)

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	cosine
Epochs	3
Max Train Steps	999
Batch Size	64
Weight Decay	0.05
Seed	7
Random Crop	False
Random Flip	False

Performance

Metric	Value
Train Accuracy	0.9986
Val Accuracy	0.9576
Test Accuracy	0.9546

Training Categories

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

turtle, aquarium_fish, table, lizard, bed, house, cattle, cloud, telephone, cockroach, chair, couch, apple, elephant, worm, squirrel, pine_tree, bridge, sea, bottle, camel, dinosaur, oak_tree, lion, seal, possum, bowl, kangaroo, dolphin, trout, sweet_pepper, shark, clock, woman, skyscraper, sunflower, caterpillar, mushroom, wolf, wardrobe, flatfish, rose, bear, whale, lawn_mower, cup, bicycle, ray, lobster, road