Model-J: MAE Model (model_idx_0272)
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 | val |
| Base Model | facebook/vit-mae-base |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|
| Learning Rate | 9e-05 |
| LR Scheduler | cosine |
| Epochs | 2 |
| Max Train Steps | 666 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 272 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9420 |
| Val Accuracy | 0.8805 |
| Test Accuracy | 0.8814 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
pine_tree, lamp, crab, trout, wolf, pear, possum, flatfish, lawn_mower, snail, mountain, apple, shark, keyboard, otter, television, oak_tree, worm, man, bear, bicycle, mouse, ray, kangaroo, bed, aquarium_fish, table, cup, camel, mushroom, tulip, elephant, train, telephone, bee, lizard, hamster, skunk, bus, dolphin, can, chair, streetcar, house, tiger, sea, motorcycle, fox, girl, leopard
