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 | constant_with_warmup |
| Epochs | 2 |
| Max Train Steps | 666 |
| Batch Size | 64 |
| Weight Decay | 0.01 |
| Seed | 790 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.2053 |
| Val Accuracy | 0.2037 |
| Test Accuracy | 0.1986 |
The model was fine-tuned on the following 50 CIFAR100 classes:
road, beetle, turtle, bicycle, wolf, plate, castle, hamster, sweet_pepper, plain, sea, palm_tree, cockroach, willow_tree, boy, lizard, fox, butterfly, beaver, snail, girl, oak_tree, spider, tiger, keyboard, can, streetcar, bus, poppy, elephant, camel, possum, kangaroo, clock, man, chimpanzee, leopard, tulip, orange, otter, bee, dinosaur, house, worm, sunflower, seal, ray, cup, wardrobe, pine_tree
Base model
facebook/dino-vitb16