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_0819)
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 | 5e-05 |
| LR Scheduler | linear |
| Epochs | 8 |
| Max Train Steps | 2664 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 819 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9983 |
| Val Accuracy | 0.8792 |
| Test Accuracy | 0.8856 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
shrew, cloud, tank, chair, girl, raccoon, willow_tree, palm_tree, squirrel, snake, rabbit, bridge, lawn_mower, plain, hamster, ray, snail, seal, house, pear, forest, worm, possum, porcupine, sunflower, bicycle, skunk, bed, otter, trout, train, rose, lamp, dinosaur, leopard, chimpanzee, boy, woman, camel, lobster, castle, maple_tree, dolphin, cup, tulip, television, clock, wolf, flatfish, can