🧼 Cleanlab Issue Handler for Tabular Datasets - Integrated for Galaxy

This repository provides a tool that detects and optionally removes or corrects label and data quality issues in tabular datasets using the Cleanlab package. It is designed to integrate with the Galaxy Project, allowing users to improve their data quality before running any downstream machine learning or statistical analysis.

---

✨ Key Features

• Supports both classification and regression tasks
• Detects:
  • 🏷️ Noisy labels
  • 🧭 Outliers
  • 🧬 Near-duplicate samples
  • 🔁 Non-i.i.d. data points

---

🧠 What is Cleanlab?

Cleanlab is an open-source Python library for:

• Data-centric AI and quality control
• Automatically identifying label issues, outliers, and duplicates
• Supporting classification and regression, across tabular, image, text, and audio data

---

📂 Repository Structure

.
├── tool/                     # Wrapped Galaxy tool (CLI + XML interface)
│   ├── cleanlab_issue_handler.py
│   └── cleanlab_issue_handler.xml
├── cleanlab_evaluators.py            # Core logic for training and evaluation
├── cleanlab_evaluation.ipynb         # Use case and evaluation of cleanning impact on different models
├── data/                    
|   ├── training_log.csv         # Stores evaluation logs
├── README.md