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 | 0.0003 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 8 |
| Max Train Steps | 2664 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 343 |
| Random Crop | True |
| Random Flip | True |
| Metric | Value |
|---|---|
| Train Accuracy | 0.5744 |
| Val Accuracy | 0.4472 |
| Test Accuracy | 0.4478 |
The model was fine-tuned on the following 50 CIFAR100 classes:
telephone, lizard, raccoon, poppy, cattle, table, skunk, caterpillar, snake, woman, chimpanzee, turtle, crab, tractor, chair, crocodile, willow_tree, otter, tank, couch, snail, lamp, bicycle, kangaroo, man, apple, sweet_pepper, streetcar, mountain, plain, tiger, elephant, whale, butterfly, maple_tree, mouse, bus, wolf, girl, sunflower, mushroom, lion, spider, porcupine, pickup_truck, rocket, orange, house, cockroach, cup
Base model
facebook/dino-vitb16