metadata
base_model: facebook/dino-vitb16
library_name: transformers
pipeline_tag: image-classification
tags:
- probex
- model-j
- weight-space-learning
Model-J: DINO Model (model_idx_0673)
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
🌐 Project | 📃 Paper | 💻 GitHub | 🤗 Dataset
Model Details
| Attribute | Value |
|---|---|
| Subset | DINO |
| Split | train |
| Base Model | facebook/dino-vitb16 |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| 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 | 673 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9998 |
| Val Accuracy | 0.9176 |
| Test Accuracy | 0.9284 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
apple, oak_tree, shark, cattle, snake, squirrel, crocodile, fox, caterpillar, seal, cup, man, dolphin, pickup_truck, cloud, palm_tree, boy, lizard, baby, streetcar, orchid, chimpanzee, aquarium_fish, dinosaur, tractor, lawn_mower, road, tulip, bottle, rabbit, worm, kangaroo, bear, beetle, turtle, snail, willow_tree, plain, porcupine, crab, clock, skunk, beaver, chair, spider, skyscraper, mushroom, mouse, train, keyboard