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 | 0.0003 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 504 |
| Random Crop | True |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.6757 |
| Val Accuracy | 0.4891 |
| Test Accuracy | 0.4938 |
The model was fine-tuned on the following 50 CIFAR100 classes:
beaver, bottle, table, can, skyscraper, trout, bowl, streetcar, shark, man, mountain, woman, willow_tree, camel, road, ray, girl, television, cockroach, leopard, lawn_mower, crab, rabbit, mouse, train, sea, snail, keyboard, butterfly, bicycle, lion, lobster, aquarium_fish, cloud, motorcycle, cup, chair, apple, dinosaur, tractor, shrew, maple_tree, bus, sunflower, plain, chimpanzee, lizard, mushroom, oak_tree, squirrel
Base model
facebook/dino-vitb16