ProbeX/Model-J__ResNet__model_idx_0009

Model-J: ResNet Model (model_idx_0009)

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	constant
Epochs	5
Max Train Steps	1665
Batch Size	64
Weight Decay	0.009
Seed	9
Random Crop	True
Random Flip	False

Performance

Metric	Value
Train Accuracy	0.9506
Val Accuracy	0.8827
Test Accuracy	0.8824

Training Categories

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

plate, television, wolf, crab, whale, skunk, telephone, couch, sea, mountain, lizard, pickup_truck, otter, rocket, cattle, kangaroo, road, cockroach, can, shark, trout, snake, bicycle, willow_tree, dinosaur, plain, bear, chimpanzee, pear, turtle, hamster, woman, ray, worm, tractor, cup, bottle, bee, lion, clock, lobster, mushroom, rabbit, seal, girl, dolphin, squirrel, keyboard, castle, apple