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 | 0.0005 |
| LR Scheduler | linear |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 320 |
| Random Crop | True |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.4385 |
| Val Accuracy | 0.3896 |
| Test Accuracy | 0.3940 |
The model was fine-tuned on the following 50 CIFAR100 classes:
orchid, lamp, bed, oak_tree, elephant, tiger, wolf, aquarium_fish, tulip, plate, cup, bottle, forest, orange, snail, chair, train, flatfish, skunk, butterfly, porcupine, leopard, squirrel, camel, crocodile, cattle, pine_tree, clock, dolphin, beaver, poppy, rocket, beetle, snake, tractor, apple, plain, bowl, lawn_mower, rose, wardrobe, whale, hamster, raccoon, caterpillar, bee, baby, bicycle, mushroom, trout
Base model
facebook/dino-vitb16