Model-J: SupViT Model (model_idx_0093)
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 | 0.0003 |
| LR Scheduler | cosine |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.05 |
| Seed | 93 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9997 |
| Val Accuracy | 0.9219 |
| Test Accuracy | 0.9244 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
clock, plate, bear, whale, maple_tree, cockroach, kangaroo, man, cloud, elephant, baby, lobster, snake, fox, pine_tree, skunk, rose, willow_tree, television, oak_tree, lawn_mower, leopard, streetcar, castle, shark, cattle, pickup_truck, house, wolf, tractor, bottle, road, sweet_pepper, pear, worm, tank, apple, chair, rabbit, lamp, rocket, flatfish, sunflower, plain, bowl, dinosaur, can, camel, tulip, lizard
