Model-J: MAE Model (model_idx_0346)
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_with_warmup |
| Epochs | 8 |
| Max Train Steps | 2664 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 346 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9939 |
| Val Accuracy | 0.8621 |
| Test Accuracy | 0.8544 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
oak_tree, boy, wolf, lobster, bottle, possum, crocodile, couch, trout, aquarium_fish, orange, plate, mouse, dolphin, caterpillar, spider, cup, pine_tree, tulip, pickup_truck, man, bowl, apple, dinosaur, bicycle, woman, seal, butterfly, lamp, castle, ray, chimpanzee, beetle, turtle, flatfish, hamster, whale, bridge, can, worm, tiger, palm_tree, sunflower, cockroach, skyscraper, lizard, house, willow_tree, beaver, keyboard
