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 | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 307 |
| Random Crop | True |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9780 |
| Val Accuracy | 0.8861 |
| Test Accuracy | 0.8854 |
The model was fine-tuned on the following 50 CIFAR100 classes:
couch, boy, crocodile, castle, wardrobe, bear, streetcar, girl, chimpanzee, keyboard, caterpillar, man, palm_tree, motorcycle, clock, bus, rocket, forest, turtle, kangaroo, ray, sea, shrew, pine_tree, lion, trout, beaver, aquarium_fish, spider, table, shark, road, dinosaur, mushroom, tank, cloud, lobster, dolphin, raccoon, skunk, fox, chair, crab, sweet_pepper, bicycle, pickup_truck, squirrel, cockroach, television, wolf
Base model
facebook/dino-vitb16