Model-J: SupViT Model (model_idx_0313)
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 | val |
| Base Model | google/vit-base-patch16-224 |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|
| Learning Rate | 0.0005 |
| LR Scheduler | cosine |
| Epochs | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 313 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9966 |
| Val Accuracy | 0.9328 |
| Test Accuracy | 0.9280 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
couch, lobster, road, motorcycle, hamster, bridge, train, seal, man, bear, bottle, cockroach, dolphin, snail, pear, forest, crocodile, can, rabbit, whale, leopard, clock, wardrobe, shark, apple, spider, lawn_mower, crab, tank, sunflower, palm_tree, house, skunk, bicycle, lamp, chimpanzee, skyscraper, tulip, sweet_pepper, cattle, porcupine, rose, camel, poppy, beetle, chair, plate, beaver, orchid, ray
