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_0927)
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 | 0.0005 |
| LR Scheduler | cosine |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 927 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.5909 |
| Val Accuracy | 0.4755 |
| Test Accuracy | 0.4642 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
camel, sweet_pepper, squirrel, poppy, bottle, cloud, otter, elephant, lion, boy, possum, television, willow_tree, apple, cup, house, couch, orange, bear, pear, woman, kangaroo, man, raccoon, bowl, aquarium_fish, palm_tree, dinosaur, motorcycle, lizard, bridge, baby, sunflower, chair, snail, crab, forest, plain, seal, lawn_mower, snake, lobster, sea, mountain, rabbit, streetcar, beetle, rose, spider, cattle