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 | 5e-05 |
| LR Scheduler | constant_with_warmup |
| Epochs | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 375 |
| Random Crop | True |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9682 |
| Val Accuracy | 0.8632 |
| Test Accuracy | 0.8674 |
The model was fine-tuned on the following 50 CIFAR100 classes:
cattle, road, clock, cup, sunflower, shrew, willow_tree, tank, pickup_truck, mountain, snake, bridge, dinosaur, bottle, maple_tree, kangaroo, leopard, skunk, butterfly, rocket, beaver, otter, girl, mushroom, plain, trout, lamp, cockroach, bed, boy, turtle, tiger, pine_tree, tulip, lion, woman, skyscraper, couch, possum, dolphin, caterpillar, bus, worm, bear, plate, lawn_mower, chair, oak_tree, streetcar, shark
Base model
facebook/dino-vitb16