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 | val |
| Base Model | facebook/dino-vitb16 |
| Dataset | CIFAR100 (50 classes) |
| Parameter | Value |
|---|---|
| Learning Rate | 7e-05 |
| LR Scheduler | cosine |
| Epochs | 7 |
| Max Train Steps | 2331 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 783 |
| Random Crop | True |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9997 |
| Val Accuracy | 0.9128 |
| Test Accuracy | 0.9136 |
The model was fine-tuned on the following 50 CIFAR100 classes:
sweet_pepper, tractor, cockroach, lion, lawn_mower, palm_tree, flatfish, shark, baby, orange, bridge, house, squirrel, orchid, snail, tank, porcupine, tulip, aquarium_fish, mushroom, man, mouse, apple, girl, clock, rose, whale, fox, otter, castle, bed, telephone, butterfly, seal, raccoon, bowl, boy, chair, tiger, trout, elephant, leopard, spider, couch, cattle, pear, oak_tree, cup, beaver, train
Base model
facebook/dino-vitb16