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 | 5e-05 |
| LR Scheduler | constant |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 319 |
| Random Crop | False |
| Random Flip | False |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9927 |
| Val Accuracy | 0.8859 |
| Test Accuracy | 0.8788 |
The model was fine-tuned on the following 50 CIFAR100 classes:
keyboard, bear, plain, bee, elephant, oak_tree, bottle, kangaroo, wardrobe, skyscraper, lizard, table, boy, castle, mouse, possum, telephone, orchid, willow_tree, bridge, seal, rabbit, bus, snail, crocodile, aquarium_fish, leopard, tulip, shark, porcupine, forest, shrew, mushroom, maple_tree, beetle, cockroach, orange, raccoon, pear, plate, cattle, bed, couch, mountain, beaver, house, ray, tiger, squirrel, lion
Base model
facebook/dino-vitb16