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 | 0.0005 |
| LR Scheduler | constant_with_warmup |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 251 |
| Random Crop | True |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.3180 |
| Val Accuracy | 0.3216 |
| Test Accuracy | 0.3028 |
The model was fine-tuned on the following 50 CIFAR100 classes:
baby, beaver, ray, sunflower, turtle, lizard, bear, palm_tree, orange, chair, otter, wardrobe, mouse, bus, cockroach, road, skunk, squirrel, cup, hamster, kangaroo, aquarium_fish, rose, lobster, apple, poppy, snake, crocodile, television, snail, tank, pine_tree, plain, lamp, cloud, orchid, bee, camel, caterpillar, chimpanzee, beetle, woman, possum, dinosaur, maple_tree, leopard, train, trout, clock, keyboard
Base model
facebook/dino-vitb16