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 | 9e-05 |
| LR Scheduler | constant |
| Epochs | 7 |
| Max Train Steps | 2331 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 455 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9708 |
| Val Accuracy | 0.8581 |
| Test Accuracy | 0.8528 |
The model was fine-tuned on the following 50 CIFAR100 classes:
keyboard, trout, aquarium_fish, snail, woman, apple, lawn_mower, worm, kangaroo, man, spider, seal, dinosaur, squirrel, beetle, willow_tree, bridge, television, raccoon, bed, snake, lizard, caterpillar, cup, bicycle, clock, shrew, cloud, oak_tree, plain, chair, rabbit, sweet_pepper, tiger, poppy, skyscraper, boy, mushroom, forest, road, table, flatfish, lamp, porcupine, streetcar, whale, motorcycle, skunk, butterfly, rose
Base model
facebook/dino-vitb16