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 | linear |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 384 |
| Random Crop | False |
| Random Flip | False |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9998 |
| Val Accuracy | 0.9376 |
| Test Accuracy | 0.9400 |
The model was fine-tuned on the following 50 CIFAR100 classes:
butterfly, kangaroo, mouse, couch, crab, cloud, beetle, castle, tulip, camel, bottle, baby, tank, bear, palm_tree, bee, skyscraper, bicycle, flatfish, house, clock, television, whale, mushroom, bridge, otter, woman, chair, lion, cockroach, road, mountain, raccoon, rocket, orange, keyboard, snake, pickup_truck, bed, oak_tree, cup, train, chimpanzee, table, lizard, apple, shark, lobster, seal, hamster
Base model
facebook/dino-vitb16