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 | 3e-05 |
| LR Scheduler | cosine |
| Epochs | 8 |
| Max Train Steps | 2664 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 681 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 1.0000 |
| Val Accuracy | 0.9365 |
| Test Accuracy | 0.9266 |
The model was fine-tuned on the following 50 CIFAR100 classes:
lion, plate, plain, ray, oak_tree, dolphin, wardrobe, caterpillar, skunk, man, trout, lawn_mower, sea, bus, kangaroo, pine_tree, camel, mountain, motorcycle, willow_tree, telephone, woman, poppy, tank, snake, squirrel, chimpanzee, orchid, rabbit, girl, can, flatfish, snail, butterfly, hamster, bear, cloud, chair, sweet_pepper, palm_tree, lizard, boy, wolf, elephant, bottle, fox, pear, dinosaur, streetcar, road
Base model
facebook/dino-vitb16