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_0638)
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 | 0.0005 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 638 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.3286 |
| Val Accuracy | 0.2952 |
| Test Accuracy | 0.2986 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
bee, tank, possum, plate, skunk, lobster, snake, lamp, elephant, bridge, cup, butterfly, hamster, otter, keyboard, snail, trout, sweet_pepper, bear, train, castle, seal, road, flatfish, can, cattle, lawn_mower, table, woman, forest, man, bed, skyscraper, lizard, wardrobe, girl, streetcar, cockroach, crab, wolf, worm, willow_tree, rabbit, orchid, apple, telephone, squirrel, poppy, clock, bicycle