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_0850)
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 | 0.0005 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.01 |
| Seed | 850 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.5928 |
| Val Accuracy | 0.4864 |
| Test Accuracy | 0.4862 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
elephant, worm, can, pear, bridge, butterfly, couch, porcupine, tiger, road, beaver, tank, crab, snail, chair, keyboard, shark, house, apple, pine_tree, flatfish, tulip, willow_tree, seal, bed, otter, lizard, tractor, man, sea, castle, snake, leopard, rabbit, bowl, hamster, forest, poppy, rocket, plain, beetle, maple_tree, turtle, television, whale, spider, woman, camel, aquarium_fish, clock