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 | 0.0001 |
| LR Scheduler | constant_with_warmup |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 704 |
| Random Crop | True |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9474 |
| Val Accuracy | 0.8149 |
| Test Accuracy | 0.8092 |
The model was fine-tuned on the following 50 CIFAR100 classes:
orchid, mouse, spider, crocodile, plate, bowl, beetle, crab, willow_tree, forest, snake, television, man, lawn_mower, tulip, pine_tree, house, seal, cockroach, bridge, shrew, pear, couch, sunflower, baby, sea, whale, hamster, lizard, cup, motorcycle, plain, shark, lobster, telephone, trout, woman, bear, butterfly, palm_tree, keyboard, streetcar, girl, cloud, camel, flatfish, chimpanzee, turtle, possum, worm
Base model
facebook/dino-vitb16