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 | 5e-05 |
| LR Scheduler | constant_with_warmup |
| Epochs | 8 |
| Max Train Steps | 2664 |
| Batch Size | 64 |
| Weight Decay | 0.05 |
| Seed | 290 |
| Random Crop | False |
| Random Flip | False |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9872 |
| Val Accuracy | 0.8784 |
| Test Accuracy | 0.8770 |
The model was fine-tuned on the following 50 CIFAR100 classes:
beetle, otter, ray, beaver, cup, clock, baby, castle, oak_tree, tank, bridge, butterfly, leopard, shark, seal, road, plate, can, pine_tree, spider, dolphin, snake, rabbit, willow_tree, bee, chimpanzee, poppy, tulip, snail, flatfish, apple, skunk, pear, camel, lizard, lamp, couch, wardrobe, lawn_mower, squirrel, mouse, aquarium_fish, skyscraper, chair, sea, mushroom, cockroach, caterpillar, possum, plain
Base model
facebook/dino-vitb16