Model-J MAE
Collection
1002 items • Updated
Model-J: MAE Model (model_idx_0206)
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 |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.01 |
| Seed | 206 |
| Random Crop | True |
| Random Flip | True |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.7566 |
| Val Accuracy | 0.5301 |
| Test Accuracy | 0.5306 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
bottle, lawn_mower, butterfly, bear, tulip, wardrobe, house, bicycle, leopard, man, woman, turtle, keyboard, otter, can, shrew, ray, orchid, bridge, dolphin, sunflower, baby, bowl, worm, plate, lizard, mountain, palm_tree, poppy, possum, forest, seal, streetcar, rabbit, caterpillar, rocket, train, hamster, porcupine, bee, telephone, couch, bed, boy, pear, aquarium_fish, tank, road, cloud, sea
- Downloads last month
- -
Model tree for ProbeX/Model-J__MAE__model_idx_0206
Base model
facebook/vit-mae-base