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 | test |
| Base Model | facebook/dino-vitb16 |
| Dataset | CIFAR100 (50 classes) |
| Parameter | Value |
|---|---|
| Learning Rate | 0.0005 |
| LR Scheduler | constant |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 478 |
| Random Crop | True |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.3164 |
| Val Accuracy | 0.2957 |
| Test Accuracy | 0.2992 |
The model was fine-tuned on the following 50 CIFAR100 classes:
mountain, tiger, baby, skunk, man, streetcar, turtle, worm, cockroach, flatfish, boy, wardrobe, bottle, spider, whale, cloud, hamster, chair, snail, mouse, telephone, girl, fox, dolphin, bee, train, bridge, lamp, plain, bed, bus, willow_tree, motorcycle, mushroom, television, ray, poppy, bowl, squirrel, shrew, wolf, tulip, sweet_pepper, sunflower, rose, porcupine, woman, house, leopard, table
Base model
facebook/dino-vitb16