Model-J DINO
Collection
1002 items
โข
Updated
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
๐ Project | ๐ Paper | ๐ป GitHub | ๐ค Dataset
| Attribute | Value |
|---|---|
| Subset | DINO |
| Split | train |
| Base Model | facebook/dino-vitb16 |
| Dataset | CIFAR100 (50 classes) |
| Parameter | Value |
|---|---|
| Learning Rate | 7e-05 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 7 |
| Max Train Steps | 2331 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 60 |
| Random Crop | True |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9997 |
| Val Accuracy | 0.9304 |
| Test Accuracy | 0.9260 |
The model was fine-tuned on the following 50 CIFAR100 classes:
mountain, cloud, flatfish, lion, mouse, bed, bear, house, can, cockroach, camel, girl, bee, motorcycle, pickup_truck, caterpillar, bridge, porcupine, baby, shark, spider, rabbit, poppy, pine_tree, tulip, cattle, kangaroo, apple, telephone, bicycle, road, tank, whale, snail, butterfly, streetcar, bottle, sunflower, plain, lizard, dolphin, trout, palm_tree, beetle, pear, boy, rocket, lawn_mower, tiger, man
Base model
facebook/dino-vitb16