ProbeX/Model-J__SupViT__model_idx_0568

Model-J: SupViT Model (model_idx_0568)

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	3e-05
LR Scheduler	cosine_with_restarts
Epochs	2
Max Train Steps	666
Batch Size	64
Weight Decay	0.03
Seed	568
Random Crop	True
Random Flip	True

Performance

Metric	Value
Train Accuracy	0.9768
Val Accuracy	0.9509
Test Accuracy	0.9426

Training Categories

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

bowl, telephone, seal, ray, bee, motorcycle, boy, mushroom, shrew, television, flatfish, turtle, road, bear, orchid, streetcar, palm_tree, train, apple, rose, orange, maple_tree, worm, possum, sea, crocodile, dolphin, aquarium_fish, lobster, crab, wolf, raccoon, chimpanzee, table, kangaroo, porcupine, lawn_mower, bridge, rocket, keyboard, bottle, skyscraper, forest, can, camel, sunflower, woman, cockroach, skunk, rabbit