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 |
| Epochs | 2 |
| Max Train Steps | 666 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 862 |
| Random Crop | True |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9803 |
| Val Accuracy | 0.9184 |
| Test Accuracy | 0.9174 |
The model was fine-tuned on the following 50 CIFAR100 classes:
aquarium_fish, fox, tank, crocodile, flatfish, turtle, maple_tree, snake, lobster, man, house, castle, mouse, mushroom, cockroach, shark, raccoon, worm, television, keyboard, ray, table, elephant, bowl, willow_tree, orchid, dinosaur, skunk, telephone, caterpillar, road, girl, bed, sunflower, beaver, seal, lawn_mower, leopard, poppy, woman, lamp, cattle, wardrobe, clock, lizard, bear, cloud, motorcycle, rabbit, baby
Base model
facebook/dino-vitb16