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_0532)
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 | val |
| Base Model | facebook/dino-vitb16 |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|---|
| Learning Rate | 7e-05 |
| LR Scheduler | constant |
| Epochs | 2 |
| Max Train Steps | 666 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 532 |
| Random Crop | True |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9496 |
| Val Accuracy | 0.8627 |
| Test Accuracy | 0.8738 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
trout, oak_tree, train, cockroach, ray, plain, pine_tree, wardrobe, elephant, forest, sweet_pepper, chair, snake, raccoon, mushroom, road, television, lawn_mower, tiger, bicycle, lobster, sea, table, girl, man, crocodile, couch, shark, orange, motorcycle, bus, pear, skyscraper, rabbit, beetle, castle, tulip, fox, poppy, skunk, boy, bowl, aquarium_fish, spider, snail, bear, cloud, hamster, kangaroo, shrew