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_0996)
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 | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.01 |
| Seed | 996 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9303 |
| Val Accuracy | 0.7720 |
| Test Accuracy | 0.7802 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
bed, raccoon, man, flatfish, tractor, otter, porcupine, fox, motorcycle, snake, wolf, cattle, tank, butterfly, maple_tree, rabbit, orange, snail, boy, apple, train, orchid, crab, baby, camel, mountain, cup, bicycle, spider, cockroach, clock, caterpillar, skyscraper, pine_tree, kangaroo, bee, shrew, seal, shark, bottle, cloud, beaver, elephant, trout, plain, girl, keyboard, rose, forest, lawn_mower