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 | 0.0001 |
| LR Scheduler | constant |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 378 |
| Random Crop | False |
| Random Flip | False |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9435 |
| Val Accuracy | 0.8187 |
| Test Accuracy | 0.8236 |
The model was fine-tuned on the following 50 CIFAR100 classes:
plate, trout, maple_tree, fox, lamp, lizard, chair, cup, chimpanzee, ray, bus, dolphin, whale, mouse, palm_tree, butterfly, man, clock, girl, lobster, pear, seal, snake, bottle, sweet_pepper, spider, mushroom, bear, caterpillar, baby, oak_tree, otter, willow_tree, sea, television, raccoon, crab, tractor, mountain, skunk, orchid, beetle, cattle, worm, cockroach, cloud, squirrel, camel, possum, pine_tree
Base model
facebook/dino-vitb16