Model-J: MAE Model (model_idx_0856)
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 | cosine |
| Epochs | 7 |
| Max Train Steps | 2331 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 856 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9946 |
| Val Accuracy | 0.6800 |
| Test Accuracy | 0.6794 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
television, train, whale, mouse, snake, pear, dinosaur, apple, possum, rabbit, palm_tree, couch, tank, bridge, cloud, plate, crocodile, fox, caterpillar, beaver, bear, aquarium_fish, table, kangaroo, chimpanzee, wardrobe, snail, shark, lawn_mower, cattle, mushroom, bed, porcupine, lizard, turtle, willow_tree, orchid, lion, plain, bus, rocket, tractor, can, girl, mountain, keyboard, hamster, house, camel, elephant
