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_0481)
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 | 0.0005 |
| LR Scheduler | constant |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.05 |
| Seed | 481 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.5780 |
| Val Accuracy | 0.4869 |
| Test Accuracy | 0.4728 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
poppy, lamp, otter, worm, leopard, chair, snake, sweet_pepper, pickup_truck, bed, crab, beaver, shrew, mountain, plate, apple, caterpillar, bowl, cattle, skunk, tulip, skyscraper, mouse, train, camel, maple_tree, beetle, rabbit, motorcycle, willow_tree, porcupine, bottle, wolf, bicycle, pear, spider, tank, shark, hamster, girl, clock, butterfly, lizard, television, whale, turtle, telephone, man, boy, wardrobe