ProbeX/Model-J__ResNet__model_idx_0000

Model-J: ResNet Model (model_idx_0000)

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	3e-05
LR Scheduler	constant
Epochs	3
Max Train Steps	999
Batch Size	64
Weight Decay	0.01
Seed	0
Random Crop	False
Random Flip	False

Performance

Metric	Value
Train Accuracy	0.7781
Val Accuracy	0.7715
Test Accuracy	0.7464

Training Categories

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

cattle, sweet_pepper, bee, dinosaur, trout, beaver, plate, raccoon, tulip, sunflower, pine_tree, apple, squirrel, spider, hamster, tractor, seal, pickup_truck, bowl, orchid, man, bed, camel, boy, dolphin, sea, oak_tree, elephant, butterfly, mountain, keyboard, maple_tree, snake, woman, cup, castle, cloud, television, rocket, motorcycle, crocodile, pear, shrew, flatfish, poppy, rabbit, lizard, orange, caterpillar, crab