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_0414)
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 | cosine_with_restarts |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 414 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9955 |
| Val Accuracy | 0.8245 |
| Test Accuracy | 0.8196 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
cup, orchid, plain, man, boy, otter, pickup_truck, chair, mushroom, poppy, lamp, sunflower, pine_tree, elephant, lizard, tulip, crocodile, lawn_mower, bridge, sea, butterfly, wolf, lobster, snail, pear, apple, raccoon, train, crab, bowl, television, squirrel, cockroach, kangaroo, bed, sweet_pepper, leopard, house, baby, dinosaur, road, table, possum, castle, telephone, trout, tiger, caterpillar, wardrobe, oak_tree