Model-J: MAE Model (model_idx_0219)
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 | 9e-05 |
| LR Scheduler | linear |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.05 |
| Seed | 219 |
| Random Crop | True |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9741 |
| Val Accuracy | 0.8821 |
| Test Accuracy | 0.8872 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
seal, forest, pear, dinosaur, television, kangaroo, camel, bear, tiger, couch, elephant, boy, porcupine, keyboard, chair, turtle, crocodile, girl, shrew, cockroach, hamster, worm, spider, road, palm_tree, skyscraper, snail, butterfly, orchid, man, oak_tree, crab, clock, mountain, baby, lobster, cup, possum, dolphin, cloud, rose, telephone, plain, squirrel, streetcar, beetle, chimpanzee, wolf, flatfish, plate
