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 | 3e-05 |
| LR Scheduler | constant_with_warmup |
| Epochs | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.05 |
| Seed | 344 |
| Random Crop | False |
| Random Flip | False |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9843 |
| Val Accuracy | 0.8920 |
| Test Accuracy | 0.8982 |
The model was fine-tuned on the following 50 CIFAR100 classes:
table, beetle, chimpanzee, lizard, lamp, wardrobe, lion, cup, leopard, bowl, mountain, beaver, tiger, baby, shrew, palm_tree, flatfish, aquarium_fish, tulip, sweet_pepper, keyboard, snake, man, bear, squirrel, poppy, crab, sea, cattle, girl, ray, mushroom, television, chair, woman, apple, rabbit, dinosaur, shark, elephant, turtle, camel, pickup_truck, kangaroo, boy, rocket, maple_tree, plain, rose, fox
Base model
facebook/dino-vitb16