Model-J: MAE Model (model_idx_0316)
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 | train |
| Base Model | facebook/vit-mae-base |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|
| Learning Rate | 0.0005 |
| LR Scheduler | linear |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 316 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.8776 |
| Val Accuracy | 0.5856 |
| Test Accuracy | 0.5778 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
dolphin, crocodile, poppy, streetcar, cup, couch, mouse, lamp, camel, tractor, crab, keyboard, motorcycle, willow_tree, lizard, bus, sunflower, bowl, rocket, pine_tree, lawn_mower, lobster, wolf, mushroom, porcupine, aquarium_fish, television, can, trout, pickup_truck, girl, bridge, fox, caterpillar, tulip, castle, skyscraper, possum, dinosaur, beaver, raccoon, apple, palm_tree, tank, plate, elephant, rose, man, worm, hamster
