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_0827)
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 | 7e-05 |
| LR Scheduler | constant_with_warmup |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.01 |
| Seed | 827 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9690 |
| Val Accuracy | 0.8755 |
| Test Accuracy | 0.8666 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
man, butterfly, cockroach, train, plate, skyscraper, raccoon, kangaroo, crocodile, otter, maple_tree, apple, bed, possum, cloud, spider, pine_tree, mouse, squirrel, woman, cup, dinosaur, bear, pickup_truck, beaver, lawn_mower, seal, castle, dolphin, sea, clock, can, porcupine, aquarium_fish, snail, ray, chair, whale, skunk, bicycle, tulip, mushroom, television, orchid, worm, wolf, couch, elephant, bee, bowl