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 | 5e-05 |
| LR Scheduler | constant_with_warmup |
| Epochs | 7 |
| Max Train Steps | 2331 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 995 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9681 |
| Val Accuracy | 0.8867 |
| Test Accuracy | 0.8840 |
The model was fine-tuned on the following 50 CIFAR100 classes:
can, telephone, lobster, motorcycle, whale, tank, bus, camel, bee, crocodile, man, beetle, pine_tree, lawn_mower, maple_tree, boy, tulip, dinosaur, leopard, hamster, lizard, crab, castle, turtle, girl, bear, fox, cup, train, cockroach, palm_tree, skyscraper, road, cloud, otter, table, forest, lion, bowl, apple, chimpanzee, bottle, orange, snake, shark, spider, orchid, television, squirrel, pear
Base model
facebook/dino-vitb16