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_0604)
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 | 7e-05 |
| LR Scheduler | constant |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 604 |
| Random Crop | True |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9748 |
| Val Accuracy | 0.8736 |
| Test Accuracy | 0.8678 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
snail, lamp, dolphin, sea, bed, rocket, man, hamster, shrew, tank, mountain, trout, skunk, cattle, beaver, lawn_mower, shark, train, dinosaur, otter, snake, elephant, possum, tractor, caterpillar, clock, whale, table, crocodile, apple, bear, crab, worm, rabbit, pickup_truck, can, lion, plain, bowl, sunflower, camel, bottle, willow_tree, baby, bus, house, tiger, porcupine, flatfish, sweet_pepper