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 | 0.0001 |
| LR Scheduler | linear |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 285 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9991 |
| Val Accuracy | 0.9053 |
| Test Accuracy | 0.9212 |
The model was fine-tuned on the following 50 CIFAR100 classes:
rabbit, table, lobster, snake, worm, crocodile, girl, caterpillar, hamster, shark, skunk, elephant, mushroom, pickup_truck, oak_tree, castle, lion, camel, crab, streetcar, bicycle, rose, train, dinosaur, bridge, bus, cockroach, orange, sunflower, dolphin, cattle, porcupine, lamp, possum, chair, poppy, mouse, clock, pear, man, mountain, flatfish, ray, cup, house, seal, fox, maple_tree, leopard, bee
Base model
facebook/dino-vitb16