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_0483)
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 | cosine_with_restarts |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 483 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9291 |
| Val Accuracy | 0.8683 |
| Test Accuracy | 0.8580 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
tulip, keyboard, lawn_mower, mouse, shark, table, chair, sunflower, caterpillar, cockroach, bear, sweet_pepper, cattle, lobster, mountain, television, can, beaver, clock, baby, raccoon, elephant, plate, bicycle, dolphin, maple_tree, kangaroo, leopard, seal, flatfish, orange, castle, bottle, bed, streetcar, bee, turtle, mushroom, palm_tree, pear, worm, rocket, lion, ray, wardrobe, orchid, forest, couch, plain, bridge