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_0219)
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 | cosine_with_restarts |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 219 |
| Random Crop | True |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.6216 |
| Val Accuracy | 0.4453 |
| Test Accuracy | 0.4574 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
dolphin, bee, poppy, mouse, possum, otter, chair, tractor, lizard, worm, plate, clock, rabbit, castle, aquarium_fish, orange, shrew, lobster, tank, leopard, house, squirrel, snail, willow_tree, bear, trout, bowl, turtle, rocket, caterpillar, cockroach, oak_tree, man, pine_tree, skunk, porcupine, bottle, cloud, can, pickup_truck, boy, lamp, spider, skyscraper, crocodile, mushroom, road, maple_tree, whale, flatfish