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_0808)
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 | 3e-05 |
| LR Scheduler | constant |
| Epochs | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 808 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9112 |
| Val Accuracy | 0.8245 |
| Test Accuracy | 0.8346 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
cockroach, telephone, chimpanzee, apple, cup, leopard, tank, camel, bear, turtle, streetcar, cattle, man, possum, road, mountain, crocodile, shark, otter, skunk, rose, skyscraper, plate, keyboard, chair, raccoon, tulip, girl, train, spider, bridge, lamp, whale, woman, bowl, lawn_mower, couch, kangaroo, pine_tree, rocket, plain, sea, orchid, boy, mushroom, porcupine, orange, can, squirrel, seal