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_0134)
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 | 0.0003 |
| LR Scheduler | linear |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 134 |
| Random Crop | True |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.4845 |
| Val Accuracy | 0.4195 |
| Test Accuracy | 0.4288 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
bed, skyscraper, clock, sea, shrew, snail, camel, castle, lamp, cattle, forest, streetcar, leopard, orchid, lizard, mouse, apple, pear, baby, porcupine, motorcycle, orange, snake, raccoon, worm, mushroom, bus, bridge, oak_tree, man, lawn_mower, maple_tree, butterfly, fox, crocodile, keyboard, mountain, bee, seal, cloud, rabbit, house, road, table, cockroach, sweet_pepper, sunflower, bear, plate, train