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 | constant_with_warmup |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 992 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.2795 |
| Val Accuracy | 0.2584 |
| Test Accuracy | 0.2618 |
The model was fine-tuned on the following 50 CIFAR100 classes:
pear, lion, mushroom, hamster, plate, table, forest, bottle, otter, snake, dolphin, squirrel, chimpanzee, possum, bus, tiger, tulip, porcupine, worm, can, spider, bridge, flatfish, bed, kangaroo, keyboard, orange, pickup_truck, mountain, rocket, shrew, road, man, crocodile, butterfly, cockroach, crab, beaver, wolf, boy, lawn_mower, raccoon, bowl, seal, oak_tree, streetcar, television, bee, apple, elephant
Base model
facebook/dino-vitb16