Model-J: SupViT Model (model_idx_0730)
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 | 3e-05 |
| LR Scheduler | constant |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 730 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9976 |
| Val Accuracy | 0.9307 |
| Test Accuracy | 0.9386 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
bee, television, castle, rose, sunflower, caterpillar, mouse, otter, willow_tree, raccoon, skunk, table, crab, apple, bowl, boy, kangaroo, porcupine, beetle, flatfish, dinosaur, squirrel, beaver, elephant, tiger, lawn_mower, wolf, tulip, sea, snail, clock, poppy, tank, butterfly, pine_tree, bed, baby, bus, ray, train, aquarium_fish, lamp, wardrobe, girl, plate, hamster, whale, fox, rabbit, oak_tree
