ProbeX/Model-J__DINO__model_idx_0004

Model-J: DINO 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	DINO
Split	train
Base Model	facebook/dino-vitb16
Dataset	CIFAR100 (50 classes)

Training Hyperparameters

Parameter	Value
Learning Rate	0.0003
LR Scheduler	linear
Epochs	6
Max Train Steps	1998
Batch Size	64
Weight Decay	0.05
Seed	4
Random Crop	True
Random Flip	True

Performance

Metric	Value
Train Accuracy	0.6470
Val Accuracy	0.5216
Test Accuracy	0.5240

Training Categories

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

pear, telephone, crab, squirrel, skyscraper, tulip, sweet_pepper, poppy, worm, skunk, keyboard, pine_tree, bus, train, lamp, castle, sea, couch, trout, forest, orange, hamster, chair, aquarium_fish, sunflower, maple_tree, streetcar, chimpanzee, lizard, bicycle, beaver, girl, palm_tree, rocket, lawn_mower, bee, baby, plain, turtle, wardrobe, wolf, porcupine, oak_tree, television, tiger, table, plate, orchid, motorcycle, house