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_0439)
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 | val |
| Base Model | facebook/vit-mae-base |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|---|
| Learning Rate | 0.0003 |
| LR Scheduler | linear |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 439 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9990 |
| Val Accuracy | 0.8619 |
| Test Accuracy | 0.8606 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
rocket, orchid, bottle, crocodile, baby, orange, crab, willow_tree, lizard, leopard, turtle, mountain, television, forest, couch, spider, chair, boy, tiger, cup, apple, shark, telephone, bus, palm_tree, poppy, seal, possum, ray, mushroom, hamster, lobster, porcupine, trout, whale, streetcar, pine_tree, cloud, bee, skunk, tractor, house, castle, lion, clock, train, rose, caterpillar, motorcycle, elephant