README.md

---
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_0933)

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

<p align="center">
    🌐 <a href="https://horwitz.ai/probex" target="_blank">Project</a> | 📃 <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | 💻 <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | 🤗 <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
</p>

![ProbeX](https://raw.githubusercontent.com/eliahuhorwitz/ProbeX/main/imgs/poster.png)

## Model Details

| Attribute | Value |
|---|---|
| **Subset** | ResNet |
| **Split** | val |
| **Base Model** | `microsoft/resnet-101` |
| **Dataset** | CIFAR100 (50 classes) |

## Training Hyperparameters

| Parameter | Value |
|---|---|
| Learning Rate | 9e-05 |
| LR Scheduler | cosine |
| Epochs | 2 |
| Max Train Steps | 666 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 933 |
| Random Crop | False |
| Random Flip | False |

## Performance

| Metric | Value |
|---|---|
| Train Accuracy | 0.8196 |
| Val Accuracy | 0.7909 |
| Test Accuracy | 0.7900 |

## Training Categories

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

`bowl`, `table`, `boy`, `bee`, `skyscraper`, `lobster`, `mushroom`, `tractor`, `bicycle`, `road`, `kangaroo`, `sunflower`, `worm`, `poppy`, `mouse`, `beetle`, `man`, `oak_tree`, `seal`, `cattle`, `raccoon`, `hamster`, `whale`, `snail`, `lamp`, `apple`, `motorcycle`, `caterpillar`, `pine_tree`, `dinosaur`, `cup`, `turtle`, `streetcar`, `pear`, `tulip`, `butterfly`, `skunk`, `chair`, `orange`, `telephone`, `beaver`, `mountain`, `palm_tree`, `shark`, `chimpanzee`, `orchid`, `pickup_truck`, `house`, `clock`, `plain`