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_0746)
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 | test |
| Base Model | facebook/dino-vitb16 |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|---|
| Learning Rate | 0.0003 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 2 |
| Max Train Steps | 666 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 746 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.3512 |
| Val Accuracy | 0.3189 |
| Test Accuracy | 0.3368 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
castle, mountain, woman, oak_tree, keyboard, rabbit, mushroom, rose, plain, bridge, butterfly, house, lamp, cup, snake, pickup_truck, shrew, girl, wolf, dolphin, orange, porcupine, chair, fox, skyscraper, seal, aquarium_fish, man, bowl, bee, kangaroo, flatfish, can, hamster, telephone, crocodile, tractor, chimpanzee, tank, shark, ray, bottle, snail, cattle, squirrel, train, motorcycle, leopard, table, spider