Skip to main content
Version: Next

Installation

Quick Start​

The fastest way to get started with xplainable is through PyPI:

pip install xplainable
Installation Complete!

That's it! You now have the core xplainable package installed and ready to use for transparent machine learning.

Installation Options​

Core Package​

The core package includes all essential features for transparent machine learning:

pip install xplainable

Includes:

  • ✅ XClassifier and XRegressor models
  • ✅ Preprocessing pipeline and transformers
  • ✅ Hyperparameter optimization
  • ✅ Model explainability and visualization
  • ✅ Partitioned models and surrogate models

GUI Features​

For interactive Jupyter notebook GUIs, install with the GUI extras:

pip install xplainable[gui]

Additional features:

  • 🎯 Interactive model training interfaces
  • 📊 Visual preprocessing tools
  • 🔧 GUI-based hyperparameter tuning
  • 📈 Interactive explanations and plots

Advanced Plotting​

For enhanced visualization capabilities:

pip install xplainable[plotting]

Additional features:

  • 📊 Advanced Altair-based visualizations
  • 🎨 Custom plot themes and styling
  • 📈 Interactive explanation plots
  • 🔍 Enhanced model inspection tools

Cloud Integration​

For cloud deployment and collaboration features:

pip install xplainable-client
Cloud Package

The cloud client is a separate package that provides integration with Xplainable Cloud for model deployment, collaboration, and production management.

Cloud features:

  • ☁️ Model deployment and management
  • 👥 Team collaboration
  • 🔄 Model versioning
  • 📊 Production monitoring
  • 🔐 Secure API deployments

Complete Installation​

For all features, install both packages:

pip install xplainable[gui,plotting]
pip install xplainable-client

Environment Setup​

🐍 Python Version

Python 3.8 - 3.11

Python 3.8 recommended for GUI features due to ipywidgets compatibility.

💻 Environment

Virtual Environment

Always use virtual environments to avoid package conflicts.

Setting Up Virtual Environment​

# Create virtual environment
python -m venv xplainable-env

# Activate environment
# On Windows:
xplainable-env\Scripts\activate
# On macOS/Linux:
source xplainable-env/bin/activate

# Install xplainable
pip install xplainable[gui,plotting]
pip install xplainable-client

Using conda​

# Create conda environment
conda create -n xplainable-env python=3.8

# Activate environment
conda activate xplainable-env

# Install xplainable
pip install xplainable[gui,plotting]
pip install xplainable-client

Jupyter Notebook Setup​

Installation​

If you don't have Jupyter installed:

pip install jupyter

Widget Extensions​

For GUI features to work properly in Jupyter:

# Install and enable widget extensions
jupyter nbextension enable --py widgetsnbextension

JupyterLab Setup​

For JupyterLab users:

pip install jupyterlab
jupyter labextension install @jupyter-widgets/jupyterlab-manager

Known Issues & Solutions​

Widget Rendering Issues​

If widgets don't render properly:

# Reinstall ipywidgets
pip uninstall ipywidgets
pip install ipywidgets==7.6.5

# Clear notebook cache
jupyter notebook --clear-cache

Import Errors​

If you encounter import errors:

# Upgrade pip and reinstall
pip install --upgrade pip
pip install --force-reinstall xplainable

Verification​

Test Core Installation​

import xplainable as xp
print(f"Xplainable version: {xp.__version__}")

# Test basic functionality
from xplainable.core.models import XClassifier
model = XClassifier()
print("✅ Core installation successful!")

Test GUI Installation​

import xplainable as xp

# This should work without errors if GUI is installed
try:
    # Test GUI components
    from xplainable.gui import classifier
    print("✅ GUI installation successful!")
except ImportError as e:
    print(f"❌ GUI installation failed: {e}")
    print("Install with: pip install xplainable[gui]")

Test Cloud Client​

try:
    from xplainable_client import Client
    print("✅ Cloud client installation successful!")
except ImportError as e:
    print(f"❌ Cloud client not installed: {e}")
    print("Install with: pip install xplainable-client")

Docker Setup​

For containerized environments:

FROM python:3.8-slim

# Install system dependencies
RUN apt-get update && apt-get install -y \
    build-essential \
    && rm -rf /var/lib/apt/lists/*

# Install xplainable
RUN pip install xplainable[gui,plotting] xplainable-client

# Set working directory
WORKDIR /app

# Copy your code
COPY . .

# Expose Jupyter port
EXPOSE 8888

# Start Jupyter
CMD ["jupyter", "notebook", "--ip=0.0.0.0", "--port=8888", "--no-browser", "--allow-root"]

Troubleshooting​

Common Issues​

ModuleNotFoundError: No module named 'xplainable'

Solution:

  • Check that you're in the correct virtual environment
  • Reinstall: pip install xplainable
  • Verify installation: pip list | grep xplainable
Widgets not displaying in Jupyter

Solution:

  • Ensure you have the GUI extras: pip install xplainable[gui]
  • Install widget extensions: jupyter nbextension enable --py widgetsnbextension
  • Restart Jupyter kernel
  • Use Python 3.8 for best compatibility
Cloud client import errors

Solution:

  • Install cloud client: pip install xplainable-client
  • Check that both packages are in the same environment
  • Verify installation: pip list | grep xplainable

Next Steps​

Ready to Build?

Now that you have xplainable installed, check out our Python API documentation or jump straight into our tutorials for hands-on examples.

Quick Start Example​

import xplainable as xp
from xplainable.core.models import XClassifier

# Load sample data
data = xp.load_dataset('titanic')
X, y = data.drop('Survived', axis=1), data['Survived']

# Train a transparent model
model = XClassifier()
model.fit(X, y)

# Get explanations
model.explain()

Cloud Integration Example​

from xplainable_client import Client
import os

# Initialize cloud client
client = Client(api_key=os.environ['XP_API_KEY'])

# Deploy your model
model_id, version_id = client.create_model(
    model=model,
    model_name="My First Model",
    model_description="Transparent Titanic survival model",
    x=X,
    y=y
)

Support​

Need help with installation?

  • 📚 Documentation: Check our comprehensive guides
  • 💬 Community: Join our user community
  • 🐛 Issues: Report bugs on GitHub
  • 📧 Enterprise: Contact us for enterprise support