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_0575)
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 | 3e-05 |
| LR Scheduler | linear |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 575 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9812 |
| Val Accuracy | 0.8739 |
| Test Accuracy | 0.8712 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
pickup_truck, keyboard, hamster, cloud, wolf, clock, willow_tree, cockroach, mountain, caterpillar, aquarium_fish, tulip, maple_tree, seal, bee, lion, can, television, camel, otter, wardrobe, plate, bicycle, snail, telephone, baby, trout, crocodile, worm, crab, beaver, porcupine, castle, streetcar, bed, palm_tree, lawn_mower, raccoon, dinosaur, woman, man, rose, lizard, possum, sea, train, table, boy, spider, bottle