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 | 0.0003 |
| LR Scheduler | constant_with_warmup |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.05 |
| Seed | 325 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.4118 |
| Val Accuracy | 0.3587 |
| Test Accuracy | 0.3508 |
The model was fine-tuned on the following 50 CIFAR100 classes:
house, castle, tank, camel, otter, hamster, woman, sweet_pepper, shark, bowl, rabbit, tulip, flatfish, cockroach, dinosaur, possum, plain, cattle, can, leopard, baby, cup, couch, road, ray, man, clock, bee, raccoon, shrew, trout, sea, skyscraper, willow_tree, crab, forest, porcupine, bridge, crocodile, rose, beetle, pickup_truck, maple_tree, skunk, lawn_mower, orchid, mountain, lizard, snake, bear
Base model
facebook/dino-vitb16