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 | 3e-05 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 8 |
| Max Train Steps | 2664 |
| Batch Size | 64 |
| Weight Decay | 0.01 |
| Seed | 380 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 1.0000 |
| Val Accuracy | 0.9339 |
| Test Accuracy | 0.9242 |
The model was fine-tuned on the following 50 CIFAR100 classes:
squirrel, spider, whale, pear, bee, castle, chair, beaver, baby, leopard, bicycle, tulip, skyscraper, tractor, shrew, porcupine, dinosaur, cloud, crab, sweet_pepper, rocket, cattle, possum, worm, skunk, sunflower, caterpillar, trout, tiger, orchid, bowl, bear, mouse, dolphin, lizard, plain, poppy, willow_tree, man, woman, plate, shark, wardrobe, beetle, pine_tree, crocodile, streetcar, telephone, snail, bus
Base model
facebook/dino-vitb16