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 | 9e-05 |
| LR Scheduler | linear |
| Epochs | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 846 |
| Random Crop | False |
| Random Flip | False |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9993 |
| Val Accuracy | 0.9157 |
| Test Accuracy | 0.9164 |
The model was fine-tuned on the following 50 CIFAR100 classes:
table, lamp, bowl, road, whale, mountain, man, bicycle, maple_tree, hamster, chimpanzee, woman, television, pickup_truck, fox, streetcar, rocket, shark, bee, baby, palm_tree, squirrel, kangaroo, snail, ray, chair, bus, orchid, can, cattle, rose, plain, couch, wolf, possum, telephone, poppy, otter, train, lizard, porcupine, bear, turtle, lion, seal, orange, keyboard, boy, beetle, apple
Base model
facebook/dino-vitb16