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 | 0.0001 |
| LR Scheduler | constant |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 131 |
| Random Crop | True |
| Random Flip | False |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9484 |
| Val Accuracy | 0.8571 |
| Test Accuracy | 0.8524 |
The model was fine-tuned on the following 50 CIFAR100 classes:
lion, poppy, caterpillar, crab, sunflower, possum, plain, aquarium_fish, streetcar, porcupine, wardrobe, lamp, keyboard, pear, sweet_pepper, table, mountain, woman, leopard, orchid, house, ray, fox, plate, bed, camel, girl, wolf, kangaroo, crocodile, cup, rabbit, can, rose, lizard, castle, motorcycle, bicycle, spider, raccoon, trout, lawn_mower, pine_tree, bee, palm_tree, worm, dinosaur, bear, train, orange
Base model
facebook/dino-vitb16