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 | test |
| Base Model | facebook/dino-vitb16 |
| Dataset | CIFAR100 (50 classes) |
| Parameter | Value |
|---|---|
| Learning Rate | 9e-05 |
| LR Scheduler | constant |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.01 |
| Seed | 820 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9652 |
| Val Accuracy | 0.8693 |
| Test Accuracy | 0.8594 |
The model was fine-tuned on the following 50 CIFAR100 classes:
bed, lizard, beaver, boy, woman, motorcycle, fox, bee, seal, man, road, dolphin, castle, beetle, chair, keyboard, house, porcupine, snail, ray, rocket, sea, lobster, rose, turtle, willow_tree, couch, pine_tree, squirrel, wardrobe, tractor, orange, cattle, sunflower, bear, snake, crocodile, chimpanzee, skyscraper, plate, train, bicycle, cloud, pear, television, can, telephone, pickup_truck, crab, palm_tree
Base model
facebook/dino-vitb16