Model-J: DINO Model (model_idx_0463)
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 | 7e-05 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 463 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9997 |
| Val Accuracy | 0.9229 |
| Test Accuracy | 0.9142 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
aquarium_fish, rose, tulip, chimpanzee, tank, plain, shark, crab, forest, spider, mushroom, sweet_pepper, bicycle, oak_tree, man, porcupine, fox, sea, willow_tree, bee, bus, shrew, caterpillar, otter, lobster, rabbit, pine_tree, poppy, snail, palm_tree, seal, pickup_truck, hamster, apple, tiger, skunk, dolphin, keyboard, squirrel, clock, beaver, plate, kangaroo, flatfish, trout, train, ray, can, motorcycle, turtle
