Model-J: SupViT Model (model_idx_0205)
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 | linear |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 205 |
| Random Crop | True |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9992 |
| Val Accuracy | 0.9421 |
| Test Accuracy | 0.9396 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
can, clock, pear, motorcycle, bridge, sea, aquarium_fish, man, road, possum, leopard, fox, woman, boy, seal, castle, butterfly, bed, tractor, girl, bottle, orchid, lawn_mower, tiger, chimpanzee, orange, baby, snail, elephant, keyboard, beetle, dolphin, ray, bowl, cup, crocodile, bear, cloud, television, oak_tree, whale, porcupine, flatfish, wardrobe, palm_tree, house, cattle, skunk, lizard, streetcar
