Eliahu's picture
Add model card
0640b92 verified
---
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_0569)
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
<p align="center">
🌐 <a href="https://horwitz.ai/probex" target="_blank">Project</a> | πŸ“ƒ <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | πŸ’» <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | πŸ€— <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
</p>
![ProbeX](https://raw.githubusercontent.com/eliahuhorwitz/ProbeX/main/imgs/poster.png)
## 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 | cosine_with_restarts |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 569 |
| Random Crop | True |
| Random Flip | False |
## Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9957 |
| Val Accuracy | 0.8829 |
| Test Accuracy | 0.8926 |
## Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
`maple_tree`, `lizard`, `orange`, `crocodile`, `mushroom`, `bee`, `rabbit`, `train`, `worm`, `wolf`, `crab`, `bicycle`, `fox`, `sea`, `telephone`, `orchid`, `skyscraper`, `plain`, `poppy`, `can`, `bear`, `kangaroo`, `clock`, `baby`, `flatfish`, `mouse`, `caterpillar`, `cattle`, `camel`, `keyboard`, `tiger`, `possum`, `hamster`, `pear`, `squirrel`, `beaver`, `whale`, `pine_tree`, `wardrobe`, `aquarium_fish`, `bowl`, `cockroach`, `lion`, `tractor`, `willow_tree`, `apple`, `lamp`, `dolphin`, `streetcar`, `cup`