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_0486)
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 | 3e-05 |
| LR Scheduler | linear |
| Epochs | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 486 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9963 |
| Val Accuracy | 0.9157 |
| Test Accuracy | 0.9096 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
rose, fox, house, shark, skyscraper, woman, rocket, leopard, wolf, palm_tree, table, spider, baby, motorcycle, possum, cup, seal, sweet_pepper, man, elephant, pear, lizard, mountain, castle, apple, ray, mushroom, dolphin, lion, oak_tree, whale, girl, beetle, beaver, chimpanzee, squirrel, caterpillar, kangaroo, boy, sunflower, lamp, bus, couch, crocodile, tulip, hamster, telephone, bicycle, lobster, forest