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_0346)
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 | 5e-05 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 7 |
| Max Train Steps | 2331 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 346 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9998 |
| Val Accuracy | 0.9152 |
| Test Accuracy | 0.9206 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
palm_tree, wolf, rose, fox, keyboard, porcupine, girl, orchid, bicycle, lobster, crocodile, chair, poppy, ray, camel, willow_tree, sunflower, seal, flatfish, boy, skyscraper, crab, orange, snail, otter, tiger, mountain, possum, pine_tree, worm, bear, clock, forest, pickup_truck, chimpanzee, cockroach, wardrobe, lamp, shark, house, beaver, motorcycle, rocket, pear, turtle, beetle, snake, bee, tulip, sweet_pepper