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_0793)

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** | train |
| **Base Model** | `microsoft/resnet-101` |
| **Dataset** | CIFAR100 (50 classes) |

## Training Hyperparameters

| Parameter | Value |
|---|---|
| Learning Rate | 3e-05 |
| LR Scheduler | linear |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 793 |
| Random Crop | False |
| Random Flip | True |

## Performance

| Metric | Value |
|---|---|
| Train Accuracy | 0.7623 |
| Val Accuracy | 0.7387 |
| Test Accuracy | 0.7434 |

## Training Categories

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

`pine_tree`, `rabbit`, `couch`, `bee`, `bus`, `mountain`, `mushroom`, `shrew`, `woman`, `flatfish`, `seal`, `dolphin`, `bicycle`, `dinosaur`, `castle`, `caterpillar`, `man`, `leopard`, `raccoon`, `train`, `beaver`, `bridge`, `chair`, `whale`, `lobster`, `wolf`, `aquarium_fish`, `motorcycle`, `willow_tree`, `maple_tree`, `skyscraper`, `trout`, `boy`, `beetle`, `table`, `keyboard`, `baby`, `plain`, `tank`, `rose`, `porcupine`, `television`, `orange`, `cattle`, `fox`, `otter`, `tiger`, `cloud`, `worm`, `snake`