Model-J: SupViT Model (model_idx_0828)
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 | 0.0001 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.01 |
| Seed | 828 |
| Random Crop | True |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9981 |
| Val Accuracy | 0.9456 |
| Test Accuracy | 0.9468 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
porcupine, lawn_mower, bowl, oak_tree, keyboard, lizard, beaver, mushroom, television, palm_tree, beetle, clock, lobster, pickup_truck, butterfly, leopard, willow_tree, bed, trout, turtle, skunk, dolphin, orchid, tulip, sunflower, poppy, cup, squirrel, table, girl, telephone, dinosaur, bee, raccoon, streetcar, maple_tree, aquarium_fish, tiger, worm, wolf, bridge, fox, whale, tank, tractor, caterpillar, rabbit, pear, forest, plate
