Model-J: DINO Model (model_idx_0016)
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 | 9e-05 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 16 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9998 |
| Val Accuracy | 0.9275 |
| Test Accuracy | 0.9264 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
butterfly, leopard, bus, hamster, sea, lizard, fox, castle, plate, flatfish, can, camel, oak_tree, streetcar, train, girl, wolf, bridge, lion, rose, aquarium_fish, crab, seal, chimpanzee, shark, cockroach, otter, tractor, skyscraper, television, bee, plain, rabbit, tulip, snail, motorcycle, porcupine, wardrobe, pear, keyboard, palm_tree, lobster, bed, pine_tree, sweet_pepper, apple, orchid, shrew, baby, mountain
