Model-J: ResNet Model (model_idx_0969)
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 | 7 |
| Max Train Steps | 2331 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 969 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.8609 |
| Val Accuracy | 0.8211 |
| Test Accuracy | 0.8182 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
whale, television, porcupine, girl, elephant, shrew, couch, road, bottle, sea, cattle, crab, lion, clock, hamster, chimpanzee, wardrobe, snail, squirrel, woman, rocket, trout, bear, otter, lizard, bee, mushroom, poppy, dinosaur, raccoon, forest, caterpillar, snake, beaver, seal, wolf, beetle, motorcycle, leopard, mountain, camel, palm_tree, tank, fox, bicycle, pickup_truck, shark, tiger, castle, bridge
