Model-J: MAE Model (model_idx_0611)
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 | 3e-05 |
| LR Scheduler | constant |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 611 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9957 |
| Val Accuracy | 0.8605 |
| Test Accuracy | 0.8668 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
snail, lamp, wolf, porcupine, flatfish, crab, skyscraper, baby, bee, snake, cloud, keyboard, trout, road, sweet_pepper, mouse, raccoon, palm_tree, fox, cockroach, man, bridge, hamster, chimpanzee, shrew, train, motorcycle, pickup_truck, plate, castle, whale, kangaroo, orchid, bowl, girl, willow_tree, tiger, rose, possum, apple, squirrel, pear, pine_tree, rabbit, ray, shark, oak_tree, lobster, forest, wardrobe
