ProbeX/Model-J__MAE__model_idx_0004

Model-J: MAE Model (model_idx_0004)

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	5e-05
LR Scheduler	cosine_with_restarts
Epochs	2
Max Train Steps	666
Batch Size	64
Weight Decay	0.05
Seed	4
Random Crop	True
Random Flip	False

Performance

Metric	Value
Train Accuracy	0.8821
Val Accuracy	0.8408
Test Accuracy	0.8358

Training Categories

The model was fine-tuned on the following 50 CIFAR100 classes:

orange, lion, mountain, lobster, cup, leopard, hamster, telephone, cattle, dolphin, whale, seal, chair, television, wardrobe, shrew, bridge, house, porcupine, lamp, clock, woman, tractor, elephant, girl, table, couch, plate, orchid, fox, beaver, plain, sea, rose, boy, cockroach, palm_tree, bear, raccoon, wolf, oak_tree, bicycle, forest, apple, bus, motorcycle, road, flatfish, pear, skunk