Model-J: MAE Model (model_idx_0367)
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
ProbeXModel Details
| Attribute | Value |
|---|
| Subset | MAE |
| Split | test |
| Base Model | facebook/vit-mae-base |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|
| Learning Rate | 3e-05 |
| LR Scheduler | cosine |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 367 |
| Random Crop | True |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9802 |
| Val Accuracy | 0.8819 |
| Test Accuracy | 0.8866 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
tank, skyscraper, whale, oak_tree, sweet_pepper, chimpanzee, tulip, can, hamster, butterfly, lobster, apple, lawn_mower, train, ray, elephant, mushroom, pear, tractor, castle, leopard, bottle, sea, shrew, bridge, girl, snail, beetle, bowl, wolf, flatfish, bicycle, pine_tree, orange, baby, poppy, pickup_truck, streetcar, turtle, snake, bed, cup, woman, lion, worm, cockroach, willow_tree, plate, fox, mouse
