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_0625)
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 | 0.0003 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 4 |
| Max Train Steps | 1332 |
| Batch Size | 64 |
| Weight Decay | 0.05 |
| Seed | 625 |
| Random Crop | False |
| Random Flip | False |
Performance
| Metric | Value |
|---|---|
| Train Accuracy | 0.9846 |
| Val Accuracy | 0.9040 |
| Test Accuracy | 0.8944 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
tiger, mouse, table, camel, tractor, castle, crab, orange, beetle, bee, baby, turtle, bottle, spider, cloud, sea, can, shrew, raccoon, keyboard, whale, tulip, sweet_pepper, bed, elephant, bowl, aquarium_fish, butterfly, forest, snail, caterpillar, pickup_truck, trout, boy, road, bus, otter, man, crocodile, plate, telephone, chimpanzee, train, television, fox, hamster, lizard, flatfish, girl, kangaroo