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_0544)
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 | 3e-05 |
| LR Scheduler | linear |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.05 |
| Seed | 544 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9503 |
| Val Accuracy | 0.8536 |
| Test Accuracy | 0.8504 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
lobster, lion, bridge, tank, butterfly, willow_tree, wolf, caterpillar, oak_tree, whale, pickup_truck, clock, squirrel, porcupine, forest, shark, seal, maple_tree, plain, lizard, aquarium_fish, leopard, rabbit, table, wardrobe, streetcar, lamp, cup, bicycle, bear, castle, chimpanzee, boy, sunflower, orchid, girl, poppy, beaver, sea, rocket, skunk, cloud, shrew, train, plate, bowl, chair, rose, hamster, cattle