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_0507)
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 | 0.0003 |
| LR Scheduler | constant |
| Epochs | 8 |
| Max Train Steps | 2664 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 507 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9550 |
| Val Accuracy | 0.8173 |
| Test Accuracy | 0.8112 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
tractor, dolphin, beaver, worm, bee, dinosaur, can, cattle, television, spider, pear, squirrel, streetcar, beetle, orange, willow_tree, train, snail, mountain, baby, house, rabbit, fox, motorcycle, tulip, pine_tree, plate, woman, whale, skyscraper, bicycle, crocodile, apple, clock, rose, girl, palm_tree, lion, tiger, bed, cockroach, kangaroo, shark, wolf, rocket, oak_tree, chair, keyboard, bus, poppy