Model-J MAE
Collection
1002 items • Updated
Model-J: MAE Model (model_idx_0173)
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 | val |
| Base Model | facebook/vit-mae-base |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|---|
| Learning Rate | 7e-05 |
| LR Scheduler | constant |
| Epochs | 3 |
| Max Train Steps | 999 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 173 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9249 |
| Val Accuracy | 0.8339 |
| Test Accuracy | 0.8276 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
kangaroo, table, pine_tree, willow_tree, house, bottle, seal, can, wolf, orchid, bowl, shrew, bus, television, sea, cloud, crocodile, girl, rocket, whale, streetcar, clock, apple, pear, rabbit, trout, tulip, cattle, turtle, cockroach, lion, leopard, poppy, motorcycle, porcupine, aquarium_fish, oak_tree, shark, bridge, crab, mountain, wardrobe, bed, beaver, bear, hamster, mouse, lamp, sunflower, rose
- Downloads last month
- -
Model tree for ProbeX/Model-J__MAE__model_idx_0173
Base model
facebook/vit-mae-base