Model-J: MAE Model (model_idx_0391)
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 | 0.0003 |
| LR Scheduler | constant_with_warmup |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.05 |
| Seed | 391 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9283 |
| Val Accuracy | 0.7435 |
| Test Accuracy | 0.7484 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
wolf, motorcycle, pine_tree, poppy, lion, keyboard, skyscraper, beaver, palm_tree, clock, can, mushroom, elephant, porcupine, tiger, plain, seal, bus, butterfly, orchid, snake, raccoon, tank, woman, bear, baby, possum, crocodile, cattle, skunk, trout, flatfish, tulip, table, shrew, train, girl, dolphin, streetcar, wardrobe, road, turtle, television, bed, bee, camel, rose, dinosaur, bowl, beetle
