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_0994)
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 |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 994 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.7647 |
| Val Accuracy | 0.4829 |
| Test Accuracy | 0.4936 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
chair, leopard, seal, mountain, mushroom, whale, sweet_pepper, streetcar, crab, orange, plate, spider, tank, trout, squirrel, rocket, maple_tree, turtle, lamp, chimpanzee, castle, telephone, shrew, crocodile, rabbit, tractor, table, hamster, bottle, television, butterfly, rose, raccoon, wardrobe, snake, train, skyscraper, apple, shark, woman, bee, worm, keyboard, ray, dolphin, beetle, bowl, cockroach, sea, wolf