Model-J: DINO Model (model_idx_0129)
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 | DINO |
| Split | train |
| Base Model | facebook/dino-vitb16 |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|
| Learning Rate | 5e-05 |
| LR Scheduler | constant_with_warmup |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 129 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9837 |
| Val Accuracy | 0.8936 |
| Test Accuracy | 0.8810 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
tulip, bus, sunflower, man, bee, bicycle, bear, bottle, snail, hamster, can, snake, telephone, aquarium_fish, cup, motorcycle, bowl, whale, tank, porcupine, house, wolf, woman, baby, mountain, dolphin, kangaroo, crocodile, possum, pine_tree, maple_tree, skyscraper, tractor, plain, palm_tree, oak_tree, dinosaur, bridge, squirrel, otter, trout, boy, spider, bed, keyboard, lawn_mower, couch, chimpanzee, shrew, lamp
