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 | cosine |
| Epochs | 7 |
| Max Train Steps | 2331 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 698 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.4827 |
| Val Accuracy | 0.3869 |
| Test Accuracy | 0.3880 |
The model was fine-tuned on the following 50 CIFAR100 classes:
shark, forest, snake, bee, caterpillar, television, leopard, motorcycle, chair, lawn_mower, boy, hamster, skyscraper, orchid, lamp, squirrel, willow_tree, beaver, elephant, pine_tree, wolf, house, apple, flatfish, plain, snail, rabbit, palm_tree, wardrobe, turtle, butterfly, mouse, sweet_pepper, poppy, baby, pickup_truck, oak_tree, trout, possum, castle, bus, sea, streetcar, chimpanzee, mushroom, girl, woman, crab, couch, sunflower
Base model
facebook/dino-vitb16