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 | cosine_with_restarts |
| Epochs | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 226 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9965 |
| Val Accuracy | 0.9240 |
| Test Accuracy | 0.9252 |
The model was fine-tuned on the following 50 CIFAR100 classes:
cloud, forest, trout, poppy, oak_tree, boy, bee, cup, mushroom, tiger, butterfly, bridge, telephone, plain, cattle, sea, bowl, house, fox, orange, streetcar, willow_tree, sweet_pepper, road, pine_tree, dolphin, spider, aquarium_fish, television, bed, pickup_truck, wolf, rose, crocodile, cockroach, tractor, ray, baby, possum, lamp, beetle, squirrel, lizard, motorcycle, dinosaur, crab, tulip, girl, caterpillar, elephant
Base model
facebook/dino-vitb16