Model-J: ResNet Model (model_idx_0251)

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	ResNet
Split	train
Base Model	`microsoft/resnet-101`
Dataset	CIFAR100 (50 classes)

Training Hyperparameters

Parameter	Value
Learning Rate	7e-05
LR Scheduler	cosine
Epochs	3
Max Train Steps	999
Batch Size	64
Weight Decay	0.009
Seed	251
Random Crop	True
Random Flip	False

Performance

Metric	Value
Train Accuracy	0.7568
Val Accuracy	0.7477
Test Accuracy	0.7296

Training Categories

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

caterpillar, otter, clock, crocodile, butterfly, lobster, sweet_pepper, pine_tree, trout, dolphin, tiger, willow_tree, snail, porcupine, bus, apple, shark, palm_tree, leopard, television, crab, rabbit, table, bowl, turtle, keyboard, raccoon, squirrel, aquarium_fish, plate, man, wardrobe, rose, lizard, elephant, boy, pickup_truck, bear, cup, castle, seal, worm, house, spider, lawn_mower, chimpanzee, tank, poppy, whale, tractor