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_0147)
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 | 0.0005 |
| LR Scheduler | constant_with_warmup |
| Epochs | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 147 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.4384 |
| Val Accuracy | 0.3899 |
| Test Accuracy | 0.3918 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
chair, fox, table, willow_tree, poppy, tulip, can, sea, bus, mouse, crocodile, orange, aquarium_fish, dolphin, shrew, snail, boy, turtle, kangaroo, cockroach, flatfish, plate, house, lawn_mower, forest, lobster, clock, caterpillar, cloud, sweet_pepper, bear, butterfly, shark, baby, apple, bottle, lamp, oak_tree, ray, bed, girl, crab, chimpanzee, palm_tree, possum, bicycle, hamster, dinosaur, snake, seal