autoexplainml 3.2.1

A modular Explainable AI framework for ML, Deep Learning, and Computer Vision

🧠 AutoExplainML

A production-ready Explainable AI framework that transforms complex ML models into human-readable insights, reports, and automated project outputs.

It supports:

• Classical Machine Learning
• Deep Learning (optional)
• Computer Vision (optional)
• Automated reporting (PDF + HTML)
• CLI + API + Web UI

---

🌐 Live Demo

• 🔗 Backend API: https://autoexplainml.onrender.com
• 🔗 Frontend UI: https://autoexplainml-ui.onrender.com

---

⚡ Installation

🧩 Core Installation

pip install autoexplainml

---

🚀 Optional Feature Packs

📊 Machine Learning Stack

pip install autoexplainml[ml]

👁 Computer Vision Stack

pip install autoexplainml[cv]

🧠 Deep Learning (PyTorch)

pip install autoexplainml[dl_torch]

🤖 Deep Learning (TensorFlow)

pip install autoexplainml[dl_tf]

🔥 Full Feature Pack (Recommended)

pip install autoexplainml[full]

---

🚀 CLI Usage

📊 Basic Analysis Mode

autoexplainml model.pkl data.csv --mode analyze

📦 Full Project Mode (Auto Reports)

autoexplainml model.pkl data.csv --mode project

---

📁 Output Generated

When using --mode project:

autoexplainml_outputs/
 ├── result.json
 ├── report.html
 ├── report.pdf

---

🚀 Features

🧠 Explainability Engine

• SHAP-based feature importance
• LIME explanations
• Permutation analysis

📊 Intelligence Layer

• Data quality checks
• Fairness & bias detection
• Model reasoning insights

📦 Automation Layer

• Full ML project generation
• Auto PDF + HTML reports
• Structured JSON outputs

🌐 Interfaces

• FastAPI backend
• Streamlit frontend
• CLI tool

---

🧠 Architecture

Frontend (Streamlit)
        ↓
FastAPI Backend
        ↓
AutoExplainML Engine
        ↓
Explainability + Intelligence Layer
        ↓
Reporting System (PDF/HTML)

---

📌 Use Cases

🎓 Students

• Auto-generate ML projects
• Submit ready-made reports
• Learn explainability easily

🧠 Data Scientists

• Understand model decisions
• Debug feature impact

🏢 Industry

• Model transparency
• AI auditability

---

⚙️ Run Locally

Backend

uvicorn backend.api:app --reload

Frontend

streamlit run frontend/app.py

---

📸 Screenshots

---

🧪 Example Workflow

from autoexplainml.core.pipeline import run_pipeline

result = run_pipeline(model, X)
print(result)

---

👨‍💻 Author

Sidhant Narang

---

🔥 Why This Project Matters

AutoExplainML bridges the gap between:

• Machine Learning models
• Human understanding
• Automated reporting systems

Making AI transparent, explainable, and usable for everyone.