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 | test |
| Base Model | facebook/dino-vitb16 |
| Dataset | CIFAR100 (50 classes) |
| Parameter | Value |
|---|---|
| Learning Rate | 5e-05 |
| LR Scheduler | constant_with_warmup |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 541 |
| Random Crop | True |
| Random Flip | False |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9697 |
| Val Accuracy | 0.8995 |
| Test Accuracy | 0.8906 |
The model was fine-tuned on the following 50 CIFAR100 classes:
rabbit, crocodile, camel, spider, clock, tank, shark, pickup_truck, kangaroo, mushroom, skyscraper, couch, mouse, whale, chimpanzee, keyboard, maple_tree, butterfly, chair, rose, orange, bridge, cup, wolf, otter, pine_tree, turtle, snail, road, caterpillar, bee, lizard, telephone, woman, tractor, bus, dinosaur, motorcycle, crab, dolphin, palm_tree, elephant, mountain, snake, ray, table, trout, lion, cockroach, bowl
Base model
facebook/dino-vitb16