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 | test |
| Base Model | facebook/dino-vitb16 |
| Dataset | CIFAR100 (50 classes) |
| Parameter | Value |
|---|---|
| Learning Rate | 5e-05 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 207 |
| Random Crop | False |
| Random Flip | False |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9997 |
| Val Accuracy | 0.9173 |
| Test Accuracy | 0.9140 |
The model was fine-tuned on the following 50 CIFAR100 classes:
motorcycle, caterpillar, bed, road, orange, lizard, bridge, camel, whale, raccoon, lawn_mower, apple, otter, tractor, shrew, girl, orchid, boy, butterfly, trout, couch, kangaroo, shark, beetle, keyboard, telephone, chair, baby, cup, palm_tree, dolphin, mouse, leopard, worm, rose, man, fox, elephant, tulip, woman, sweet_pepper, oak_tree, mountain, seal, bowl, aquarium_fish, rocket, skunk, tiger, pickup_truck
Base model
facebook/dino-vitb16