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_0401)
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 | 0.0001 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.05 |
| Seed | 401 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9960 |
| Val Accuracy | 0.8816 |
| Test Accuracy | 0.8804 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
caterpillar, chair, flatfish, tractor, spider, castle, dolphin, ray, telephone, crab, plain, man, kangaroo, palm_tree, tiger, turtle, camel, boy, television, seal, apple, maple_tree, baby, rabbit, mountain, trout, motorcycle, rocket, hamster, can, leopard, cloud, keyboard, orange, sunflower, elephant, worm, lamp, bear, girl, house, clock, whale, wolf, lion, crocodile, bridge, road, plate, shark