Quantum Data Embedding Suite

A comprehensive Python package for advanced classical-to-quantum data embedding techniques designed to maximize quantum advantage in machine learning applications.

🚀 Features

• Flexible Quantum Feature Maps: Angle encoding, amplitude encoding, IQP circuits, data re-uploading, and Hamiltonian-based embeddings
• Quantum Kernel Computation: Advanced kernel methods with visualization capabilities
• Comprehensive Metrics: Expressibility, trainability, and curvature analysis
• Dimensionality Reduction: qPCA, quantum SVD, and entanglement-preserving methods
• Multi-Backend Support: Qiskit, PennyLane with real QPU compatibility (IBM, IonQ, AWS Braket)
• Benchmarking Tools: Performance evaluation across different embedding strategies
• Interactive CLI: Rapid experimentation with qdes-cli
• Extensible Architecture: Plugin support for custom ansätze and optimizers

📦 Installation

pip install quantum-data-embedding-suite

For development installation:

git clone https://github.com/krish567366/quantum-data-embedding-suite.git
cd quantum-data-embedding-suite
pip install -e ".[dev,docs]"

🎯 Quick Start

from quantum_data_embedding_suite import QuantumEmbeddingPipeline
from sklearn.datasets import load_iris
import numpy as np

# Load data
X, y = load_iris(return_X_y=True)
X = X[:50, :2]  # Use first 50 samples, 2 features

# Create embedding pipeline
pipeline = QuantumEmbeddingPipeline(
    embedding_type="angle",
    n_qubits=4,
    backend="qiskit"
)

# Embed data and compute quantum kernel
quantum_kernel = pipeline.fit_transform(X)

# Evaluate embedding quality
metrics = pipeline.evaluate_embedding(X)
print(f"Expressibility: {metrics['expressibility']:.3f}")
print(f"Trainability: {metrics['trainability']:.3f}")

🛠️ CLI Usage

# Quick benchmark on sample data
qdes-cli benchmark --dataset iris --embedding angle --n-qubits 4

# Generate embedding comparison report
qdes-cli compare --embeddings angle,amplitude,iqp --dataset wine

# Visualize quantum kernel
qdes-cli visualize --embedding angle --data my_data.csv --output kernel_plot.png

📚 Core Components

Embeddings

• AngleEmbedding: Encodes features as rotation angles
• AmplitudeEmbedding: Encodes features in quantum state amplitudes
• IQPEmbedding: Instantaneous Quantum Polynomial circuits
• DataReuploadingEmbedding: Multi-layer feature encoding
• HamiltonianEmbedding: Physics-inspired feature maps

Quantum Kernels

• Fidelity-based kernels
• Projected quantum kernels
• Trainable quantum kernels

Metrics & Analysis

• Expressibility measurement
• Gradient variance (barren plateau detection)
• Geometric curvature analysis
• Entanglement spectrum analysis

Dimensionality Reduction

• Quantum Principal Component Analysis (qPCA)
• Quantum Singular Value Decomposition
• Entanglement-preserving projections

🎓 Examples

Explore our comprehensive Jupyter notebook examples:

1. Basic Embeddings: Introduction to quantum feature maps
2. Kernel Comparison: Classical vs quantum kernel performance
3. Expressibility Analysis: Understanding embedding expressiveness
4. Real QPU Usage: Running on IBM Quantum and IonQ devices
5. Custom Ansätze: Building domain-specific embeddings

🔧 Advanced Configuration

Custom Ansatz via YAML

ansatz:
  name: "custom_variational"
  layers: 3
  entangling_gates: ["cx", "cz"]
  rotation_gates: ["rx", "ry", "rz"]
  parameter_sharing: "layer_wise"
  
optimization:
  method: "bayesian"
  acquisition: "ei"
  n_calls: 100

Backend Configuration

from quantum_data_embedding_suite.backends import IBMBackend

backend = IBMBackend(
    device="ibmq_qasm_simulator",
    shots=1024,
    optimization_level=3
)

📊 Benchmarking Results

Our benchmarking suite demonstrates quantum advantage across various datasets:

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for details.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Quantum computing frameworks: Qiskit, PennyLane
• Research inspiration from quantum machine learning literature
• Community feedback and contributions

📞 Support

• Documentation: Full Documentation
• Issues: GitHub Issues
• Discussions: GitHub Discussions

Author: Krishna Bajpai (bajpaikrishna715@gmail.com)
Maintainer: Krishna Bajpai
Version: 0.1.0