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 | 0.0005 |
| LR Scheduler | constant_with_warmup |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.01 |
| Seed | 323 |
| Random Crop | True |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.3465 |
| Val Accuracy | 0.3227 |
| Test Accuracy | 0.3214 |
The model was fine-tuned on the following 50 CIFAR100 classes:
mushroom, squirrel, dolphin, pine_tree, rose, shrew, possum, telephone, turtle, flatfish, boy, cloud, bus, tank, maple_tree, television, bed, pickup_truck, cattle, camel, woman, bridge, lawn_mower, spider, man, dinosaur, chair, skyscraper, plain, lion, couch, willow_tree, fox, mountain, shark, poppy, wolf, beetle, tractor, kangaroo, leopard, hamster, lobster, skunk, elephant, mouse, streetcar, castle, crocodile, orchid
Base model
facebook/dino-vitb16