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 | 7e-05 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 7 |
| Max Train Steps | 2331 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 735 |
| Random Crop | False |
| Random Flip | False |
| Metric | Value |
|---|---|
| Train Accuracy | 0.9999 |
| Val Accuracy | 0.9200 |
| Test Accuracy | 0.9210 |
The model was fine-tuned on the following 50 CIFAR100 classes:
bowl, hamster, chair, sweet_pepper, keyboard, motorcycle, forest, television, boy, woman, trout, camel, plate, lizard, leopard, house, dinosaur, elephant, table, beetle, kangaroo, cloud, lawn_mower, mountain, crab, willow_tree, girl, snake, sea, wolf, road, skunk, skyscraper, turtle, cockroach, possum, rabbit, tank, bus, lamp, otter, spider, caterpillar, porcupine, pine_tree, can, mushroom, wardrobe, oak_tree, poppy
Base model
facebook/dino-vitb16