Model-J: SupViT Model (model_idx_0515)
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 | test |
| Base Model | google/vit-base-patch16-224 |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|
| Learning Rate | 7e-05 |
| LR Scheduler | cosine |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 515 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9998 |
| Val Accuracy | 0.9520 |
| Test Accuracy | 0.9494 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
house, baby, cup, whale, tank, motorcycle, raccoon, bridge, dinosaur, train, bee, woman, man, streetcar, lawn_mower, shrew, table, caterpillar, sea, crab, can, cattle, lobster, tractor, willow_tree, skunk, bowl, lion, flatfish, sunflower, tiger, camel, maple_tree, orchid, bed, aquarium_fish, snake, kangaroo, squirrel, rose, cloud, mushroom, shark, mountain, wolf, bus, skyscraper, ray, chair, worm
