Model-J: SupViT Model (model_idx_0217)
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
ProbeXModel Details
| Attribute | Value |
|---|
| Subset | SupViT |
| Split | train |
| Base Model | google/vit-base-patch16-224 |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|
| Learning Rate | 7e-05 |
| LR Scheduler | constant |
| Epochs | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 217 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9887 |
| Val Accuracy | 0.9395 |
| Test Accuracy | 0.9406 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
porcupine, tulip, beaver, leopard, fox, woman, trout, rabbit, bus, house, lizard, rose, chair, tank, mountain, tractor, lawn_mower, possum, pear, worm, man, lion, plain, lobster, crab, boy, orchid, wardrobe, aquarium_fish, bear, caterpillar, skyscraper, forest, bed, turtle, telephone, tiger, butterfly, mushroom, keyboard, hamster, television, table, beetle, sunflower, cup, elephant, baby, motorcycle, cockroach
