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_0564)
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 | test |
| Base Model | facebook/vit-mae-base |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|---|
| Learning Rate | 0.0001 |
| LR Scheduler | constant_with_warmup |
| Epochs | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 564 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9603 |
| Val Accuracy | 0.8451 |
| Test Accuracy | 0.8446 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
shark, sweet_pepper, squirrel, telephone, bowl, tiger, wardrobe, palm_tree, otter, table, forest, fox, aquarium_fish, beetle, lawn_mower, skyscraper, mouse, plate, porcupine, cattle, elephant, can, bridge, raccoon, tractor, train, apple, rose, orange, butterfly, worm, leopard, snake, motorcycle, keyboard, crab, orchid, plain, bicycle, lobster, shrew, lamp, woman, seal, bus, willow_tree, turtle, caterpillar, dinosaur, castle