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_0056)
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 | 3e-05 |
| LR Scheduler | constant |
| Epochs | 7 |
| Max Train Steps | 2331 |
| Batch Size | 64 |
| Weight Decay | 0.05 |
| Seed | 56 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9925 |
| Val Accuracy | 0.8877 |
| Test Accuracy | 0.8860 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
pear, lamp, bicycle, castle, dolphin, plate, dinosaur, telephone, girl, lion, otter, kangaroo, woman, house, fox, whale, turtle, skunk, chair, tiger, squirrel, tulip, television, snake, bee, tractor, rocket, crocodile, palm_tree, mushroom, possum, caterpillar, pickup_truck, orange, table, sunflower, worm, cockroach, bridge, bottle, mouse, sea, road, streetcar, lobster, apple, pine_tree, orchid, maple_tree, train