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 | cosine |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 881 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.3760 |
| Val Accuracy | 0.3384 |
| Test Accuracy | 0.3422 |
The model was fine-tuned on the following 50 CIFAR100 classes:
pickup_truck, cloud, bridge, sunflower, keyboard, beaver, lawn_mower, bowl, ray, lizard, cockroach, tank, fox, worm, seal, bee, plate, caterpillar, pine_tree, lobster, flatfish, castle, orchid, elephant, bus, baby, squirrel, wolf, poppy, bear, rose, raccoon, cup, mountain, shrew, streetcar, porcupine, lamp, beetle, lion, trout, road, possum, sea, cattle, rabbit, motorcycle, willow_tree, chimpanzee, bed
Base model
facebook/dino-vitb16