Model-J: DINO Model (model_idx_0203)
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 | 0.0001 |
| LR Scheduler | cosine |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.01 |
| Seed | 203 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9981 |
| Val Accuracy | 0.9165 |
| Test Accuracy | 0.9186 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
lion, pear, beaver, rabbit, wardrobe, crab, mushroom, lamp, cattle, train, aquarium_fish, apple, porcupine, shrew, motorcycle, squirrel, bicycle, pine_tree, can, bee, television, keyboard, woman, chimpanzee, willow_tree, whale, orchid, shark, leopard, skyscraper, bridge, maple_tree, palm_tree, clock, skunk, streetcar, mountain, lizard, otter, flatfish, poppy, lobster, turtle, bear, hamster, tiger, trout, tractor, bowl, bus
