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_0016)
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 | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 16 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9606 |
| Val Accuracy | 0.5357 |
| Test Accuracy | 0.5438 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
pickup_truck, camel, chimpanzee, forest, lion, wardrobe, bear, raccoon, lobster, cloud, kangaroo, train, castle, streetcar, tank, lamp, spider, sunflower, lawn_mower, telephone, can, chair, caterpillar, pine_tree, bee, oak_tree, cattle, couch, road, dolphin, rabbit, television, ray, rocket, plain, cockroach, man, tiger, keyboard, beetle, fox, cup, dinosaur, squirrel, sea, bowl, apple, baby, trout, hamster