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_0754)
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 | 7e-05 |
| LR Scheduler | constant_with_warmup |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 754 |
| Random Crop | True |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9464 |
| Val Accuracy | 0.8645 |
| Test Accuracy | 0.8548 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
rabbit, tulip, rose, man, cattle, lion, elephant, bottle, butterfly, possum, tractor, rocket, beetle, motorcycle, camel, skunk, can, fox, pear, bear, chair, castle, table, turtle, bed, otter, lobster, dolphin, kangaroo, train, bus, telephone, sea, leopard, sunflower, plate, wardrobe, worm, sweet_pepper, ray, porcupine, mouse, streetcar, mountain, house, television, willow_tree, wolf, crocodile, seal