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_0334)
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 | 0.0005 |
| LR Scheduler | cosine |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 334 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.4812 |
| Val Accuracy | 0.3789 |
| Test Accuracy | 0.3970 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
pickup_truck, pear, squirrel, lion, oak_tree, telephone, flatfish, streetcar, dinosaur, skunk, clock, cloud, whale, poppy, bridge, tulip, bicycle, rabbit, train, sunflower, worm, caterpillar, chimpanzee, ray, motorcycle, beetle, rose, kangaroo, crab, table, mouse, forest, raccoon, porcupine, camel, tank, dolphin, plate, castle, boy, elephant, bowl, spider, orange, lamp, lobster, palm_tree, beaver, willow_tree, turtle