Model-J: DINO Model (model_idx_0317)
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 | val |
| Base Model | facebook/dino-vitb16 |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|
| Learning Rate | 0.0005 |
| LR Scheduler | constant |
| Epochs | 8 |
| Max Train Steps | 2664 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 317 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.4066 |
| Val Accuracy | 0.3592 |
| Test Accuracy | 0.3586 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
dolphin, skyscraper, aquarium_fish, lamp, shark, caterpillar, elephant, squirrel, whale, mushroom, table, poppy, wolf, sea, leopard, shrew, castle, tractor, motorcycle, pickup_truck, flatfish, porcupine, rabbit, tiger, worm, mouse, plain, hamster, apple, cloud, orange, camel, bear, forest, lion, bee, fox, ray, wardrobe, palm_tree, raccoon, can, seal, girl, tank, trout, snake, train, mountain, sunflower
