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 | cosine_with_restarts |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 20 |
| Random Crop | True |
| Random Flip | False |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9965 |
| Val Accuracy | 0.9056 |
| Test Accuracy | 0.9038 |
The model was fine-tuned on the following 50 CIFAR100 classes:
leopard, boy, bee, plate, cloud, mouse, camel, porcupine, cockroach, rose, oak_tree, squirrel, sunflower, apple, kangaroo, man, rocket, crab, forest, girl, pickup_truck, bus, possum, whale, bed, caterpillar, tulip, mushroom, palm_tree, maple_tree, spider, butterfly, tiger, sea, skyscraper, keyboard, motorcycle, chair, ray, rabbit, poppy, bottle, pear, wardrobe, pine_tree, clock, train, lizard, dinosaur, beetle
Base model
facebook/dino-vitb16