metadata
base_model: facebook/vit-mae-base
library_name: transformers
pipeline_tag: image-classification
tags:
- probex
- model-j
- weight-space-learning
Model-J: MAE Model (model_idx_0853)
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 | MAE |
| Split | train |
| Base Model | facebook/vit-mae-base |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|---|
| Learning Rate | 0.0005 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 8 |
| Max Train Steps | 2664 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 853 |
| Random Crop | True |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9446 |
| Val Accuracy | 0.6227 |
| Test Accuracy | 0.6414 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
television, tulip, lion, camel, castle, otter, willow_tree, sunflower, fox, beetle, orchid, spider, trout, train, mouse, poppy, seal, butterfly, bicycle, woman, tractor, chimpanzee, wardrobe, lawn_mower, turtle, road, leopard, palm_tree, rocket, raccoon, cup, keyboard, man, rose, skunk, dolphin, plate, plain, worm, caterpillar, elephant, snail, cloud, lobster, motorcycle, shrew, sweet_pepper, pine_tree, porcupine, skyscraper