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 | 0.0005 |
| LR Scheduler | constant_with_warmup |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.01 |
| Seed | 949 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.4136 |
| Val Accuracy | 0.3451 |
| Test Accuracy | 0.3542 |
The model was fine-tuned on the following 50 CIFAR100 classes:
train, skyscraper, shrew, pear, keyboard, pickup_truck, cup, fox, poppy, pine_tree, ray, telephone, sweet_pepper, spider, lion, plain, crocodile, trout, crab, cloud, wardrobe, plate, wolf, bus, beaver, couch, raccoon, lizard, house, tiger, mouse, rocket, bottle, beetle, dinosaur, caterpillar, kangaroo, mountain, butterfly, bowl, television, girl, snake, mushroom, sea, tulip, tractor, turtle, bicycle, rabbit
Base model
facebook/dino-vitb16