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_0851)
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 | 0.0005 |
| LR Scheduler | cosine |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 851 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.4475 |
| Val Accuracy | 0.3565 |
| Test Accuracy | 0.3756 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
sea, lamp, camel, dinosaur, tiger, can, bowl, turtle, apple, chimpanzee, maple_tree, whale, bee, woman, girl, oak_tree, mountain, tulip, lobster, aquarium_fish, spider, raccoon, flatfish, crocodile, television, otter, chair, cattle, castle, cup, road, pear, lion, orange, bus, bicycle, rabbit, worm, caterpillar, porcupine, bed, beaver, tractor, palm_tree, rocket, bridge, plate, ray, shark, keyboard