Model-J: ResNet Model (model_idx_0225)

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	val
Base Model	`microsoft/resnet-101`
Dataset	CIFAR100 (50 classes)

Training Hyperparameters

Parameter	Value
Learning Rate	9e-05
LR Scheduler	constant
Epochs	4
Max Train Steps	1332
Batch Size	64
Weight Decay	0.05
Seed	225
Random Crop	True
Random Flip	False

Performance

Metric	Value
Train Accuracy	0.9392
Val Accuracy	0.8619
Test Accuracy	0.8682

Training Categories

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

telephone, man, worm, porcupine, television, poppy, snail, elephant, lizard, mountain, shark, castle, bee, ray, otter, cloud, forest, crocodile, possum, plate, tractor, house, bicycle, mouse, beetle, pine_tree, pickup_truck, dolphin, palm_tree, clock, bottle, road, cup, leopard, woman, caterpillar, sweet_pepper, maple_tree, lawn_mower, cattle, fox, shrew, whale, flatfish, skunk, sea, orange, chair, kangaroo, train