metadata
base_model: microsoft/resnet-101
library_name: transformers
pipeline_tag: image-classification
tags:
- probex
- model-j
- weight-space-learning
Model-J: ResNet Model (model_idx_0311)
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 | 0.0005 |
| LR Scheduler | constant_with_warmup |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 311 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9892 |
| Val Accuracy | 0.8675 |
| Test Accuracy | 0.8722 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
kangaroo, camel, dolphin, tractor, cockroach, ray, orange, mountain, man, wolf, telephone, beetle, bridge, bee, turtle, elephant, porcupine, table, couch, flatfish, rose, rabbit, plate, maple_tree, tiger, girl, squirrel, lamp, raccoon, train, hamster, lizard, lobster, streetcar, television, whale, beaver, seal, skyscraper, dinosaur, butterfly, caterpillar, baby, otter, cup, crab, chimpanzee, road, crocodile, bear