Model-J: MAE Model (model_idx_0417)
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 | test |
| Base Model | facebook/vit-mae-base |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|
| Learning Rate | 9e-05 |
| LR Scheduler | constant_with_warmup |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 417 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9815 |
| Val Accuracy | 0.8584 |
| Test Accuracy | 0.8456 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
beaver, streetcar, snail, skunk, rocket, shark, hamster, cockroach, bed, dolphin, chair, turtle, snake, house, seal, sea, spider, crocodile, mushroom, boy, possum, castle, crab, bowl, girl, lobster, tiger, porcupine, bee, mountain, plate, telephone, clock, elephant, oak_tree, lawn_mower, skyscraper, plain, caterpillar, wardrobe, tulip, pickup_truck, fox, road, woman, couch, ray, keyboard, poppy, cattle
