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_0537)
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 | 0.0001 |
| LR Scheduler | cosine |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 537 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 1.0000 |
| Val Accuracy | 0.9283 |
| Test Accuracy | 0.9268 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
lamp, camel, orange, aquarium_fish, maple_tree, bus, table, girl, crab, cloud, plain, cockroach, telephone, baby, squirrel, spider, tiger, seal, chair, beetle, lawn_mower, snail, cup, bridge, can, caterpillar, bicycle, bee, chimpanzee, clock, wolf, bottle, possum, cattle, rose, hamster, tulip, pickup_truck, turtle, bear, television, porcupine, willow_tree, pear, whale, mouse, worm, kangaroo, apple, keyboard