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_0086)

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 | 0.0003 |
| LR Scheduler | cosine_with_restarts |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.007 |
| Seed | 86 |
| Random Crop | True |
| Random Flip | True |

## Performance

| Metric | Value |
|---|---|
| Train Accuracy | 0.9911 |
| Val Accuracy | 0.8920 |
| Test Accuracy | 0.8904 |

## Training Categories

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

`palm_tree`, `maple_tree`, `beaver`, `pine_tree`, `clock`, `bowl`, `possum`, `willow_tree`, `leopard`, `forest`, `bus`, `pickup_truck`, `tulip`, `raccoon`, `chair`, `ray`, `woman`, `couch`, `wolf`, `orchid`, `can`, `crab`, `tiger`, `lizard`, `whale`, `plate`, `butterfly`, `lion`, `caterpillar`, `beetle`, `bed`, `orange`, `mountain`, `bottle`, `otter`, `bear`, `skunk`, `television`, `rabbit`, `kangaroo`, `road`, `spider`, `train`, `trout`, `lobster`, `skyscraper`, `tractor`, `cockroach`, `dinosaur`, `oak_tree`