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 | cosine |
| Epochs | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 858 |
| Random Crop | False |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9977 |
| Val Accuracy | 0.9163 |
| Test Accuracy | 0.9104 |
The model was fine-tuned on the following 50 CIFAR100 classes:
house, table, man, chair, camel, bed, chimpanzee, beetle, train, forest, otter, sunflower, rose, butterfly, skyscraper, couch, spider, shark, bee, whale, raccoon, girl, lion, bridge, mouse, wolf, pickup_truck, cattle, rabbit, crab, tulip, clock, dolphin, orchid, tank, cockroach, leopard, bicycle, flatfish, dinosaur, beaver, possum, seal, poppy, bus, hamster, television, plate, streetcar, lobster
Base model
facebook/dino-vitb16