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_0791)
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 | 0.0003 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 791 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.4350 |
| Val Accuracy | 0.3757 |
| Test Accuracy | 0.3708 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
couch, beetle, man, pear, mushroom, rose, otter, house, flatfish, crab, skyscraper, palm_tree, shark, kangaroo, leopard, boy, baby, aquarium_fish, dolphin, mountain, snail, snake, shrew, bridge, bed, ray, motorcycle, cockroach, fox, butterfly, dinosaur, plate, skunk, seal, sweet_pepper, poppy, willow_tree, camel, bottle, orchid, sea, keyboard, forest, crocodile, tiger, table, can, beaver, train, woman