Model-J MAE
Collection
1002 items • Updated
Model-J: MAE Model (model_idx_0682)
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
🌐 Project | 📃 Paper | 💻 GitHub | 🤗 Dataset
Model Details
| Attribute | Value |
|---|---|
| Subset | MAE |
| Split | train |
| Base Model | facebook/vit-mae-base |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|---|
| Learning Rate | 7e-05 |
| LR Scheduler | constant_with_warmup |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 682 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9850 |
| Val Accuracy | 0.8872 |
| Test Accuracy | 0.8804 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
bee, squirrel, bed, kangaroo, chair, orchid, snake, poppy, leopard, couch, house, bicycle, crab, pear, trout, camel, plate, girl, rabbit, mushroom, flatfish, sunflower, lizard, aquarium_fish, raccoon, television, skyscraper, forest, cup, palm_tree, fox, butterfly, chimpanzee, snail, orange, clock, beaver, oak_tree, lion, can, turtle, lamp, crocodile, road, hamster, porcupine, cloud, wardrobe, man, apple
- Downloads last month
- 8
Model tree for ProbeX/Model-J__MAE__model_idx_0682
Base model
facebook/vit-mae-base