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_0169)
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 | constant |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 169 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9784 |
| Val Accuracy | 0.8557 |
| Test Accuracy | 0.8618 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
maple_tree, tractor, snake, sweet_pepper, seal, butterfly, bus, beaver, crab, road, poppy, keyboard, couch, otter, telephone, dinosaur, orchid, squirrel, mountain, clock, pine_tree, beetle, table, dolphin, train, lion, bridge, orange, rose, porcupine, girl, streetcar, camel, elephant, whale, shrew, bicycle, bear, wardrobe, man, tank, possum, oak_tree, crocodile, cloud, rocket, tulip, cattle, plate, snail