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 | linear |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 365 |
| Random Crop | True |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9994 |
| Val Accuracy | 0.9189 |
| Test Accuracy | 0.9176 |
The model was fine-tuned on the following 50 CIFAR100 classes:
clock, bowl, leopard, maple_tree, cattle, willow_tree, orchid, cloud, crocodile, hamster, mushroom, skyscraper, pickup_truck, house, flatfish, keyboard, television, porcupine, camel, beetle, dolphin, pear, tractor, telephone, snail, bed, aquarium_fish, plain, cockroach, crab, castle, fox, woman, ray, lizard, bus, mountain, bee, baby, rocket, apple, boy, whale, forest, road, rose, mouse, skunk, wolf, cup
Base model
facebook/dino-vitb16