Model-J: ResNet Model (model_idx_0160)
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
ProbeXModel 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 | cosine_with_restarts |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.01 |
| Seed | 160 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.7152 |
| Val Accuracy | 0.7032 |
| Test Accuracy | 0.6944 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
girl, orange, pickup_truck, sweet_pepper, sea, whale, palm_tree, butterfly, lamp, telephone, dinosaur, orchid, shrew, plate, crab, mountain, tank, forest, pine_tree, cattle, kangaroo, bottle, elephant, caterpillar, clock, train, bear, sunflower, leopard, rose, oak_tree, apple, cup, television, aquarium_fish, poppy, castle, shark, wolf, rocket, squirrel, maple_tree, turtle, seal, bee, lion, tulip, man, bowl, house
