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_0409)
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 | val |
| Base Model | facebook/vit-mae-base |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|---|
| Learning Rate | 0.0003 |
| LR Scheduler | linear |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.01 |
| Seed | 409 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.7864 |
| Val Accuracy | 0.5384 |
| Test Accuracy | 0.5406 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
fox, kangaroo, butterfly, boy, bee, otter, wardrobe, flatfish, bus, leopard, bowl, elephant, man, crocodile, ray, bottle, turtle, cloud, tractor, poppy, lawn_mower, chair, woman, dolphin, road, television, willow_tree, crab, castle, porcupine, table, train, bridge, chimpanzee, mushroom, skyscraper, apple, snail, couch, lamp, house, mountain, bed, rose, beetle, caterpillar, forest, pine_tree, beaver, aquarium_fish