Skip to content

Quantum Data Embedding Suite

PyPI - Version PyPI Downloads Python 3.8+ License: Commercial Docs

Welcome to the 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.

🎯 What is Quantum Data Embedding?

Quantum data embedding is the process of encoding classical data into quantum states, enabling quantum algorithms to process classical information. The choice of embedding significantly impacts the performance of quantum machine learning algorithms, determining:

  • Expressibility: How well the embedding can represent diverse quantum states
  • Trainability: How effectively gradients can be computed for optimization
  • Quantum Advantage: The potential for quantum speedup over classical methods

✨ Key Features

Embedding Types

  • Angle Embedding: Encode data as rotation angles in quantum gates
  • Amplitude Embedding: Encode data directly in quantum state amplitudes
  • IQP Embedding: Instantaneous Quantum Polynomial circuits for data encoding
  • Data Re-uploading: Multiple encoding layers for increased expressivity
  • Hamiltonian Embedding: Physics-inspired encodings using problem Hamiltonians

Quantum Kernels

  • Fidelity Kernels: State overlap-based similarity measures
  • Projected Kernels: Measurement-based kernel computation
  • Trainable Kernels: Parameterized kernels with gradient-based optimization

Quality Metrics

  • Expressibility: Measure how uniformly embeddings cover Hilbert space
  • Trainability: Analyze gradient magnitudes and barren plateau susceptibility
  • Gradient Variance: Evaluate optimization landscape characteristics

Advanced Features

  • Multi-Backend Support: Seamless integration with Qiskit and PennyLane
  • Real QPU Support: Execute on actual quantum hardware
  • Dimensionality Reduction: Quantum PCA and SVD implementations
  • Benchmarking Tools: Systematic classical vs quantum performance comparison
  • Interactive CLI: Rapid experimentation with qdes-cli command-line interface
  • Extensible Design: Plugin architecture for custom embeddings and metrics

🚀 Quick Start

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

# Load sample data
X, y = load_iris(return_X_y=True)
X = X[:50, :2]  # Use subset for demo

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

# Compute quantum kernel matrix
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}")

📊 Why Quantum Embeddings Matter

Quantum embeddings can provide exponential advantages in certain machine learning tasks by:

  • Exploiting Quantum Superposition: Representing data in superposition states
  • Leveraging Entanglement: Creating complex correlations between features
  • Accessing Larger Feature Spaces: Exponentially large Hilbert spaces
  • Enabling Quantum Kernels: Computing kernels that are hard to evaluate classically

🔬 Research Applications

This package supports research in:

  • Quantum Machine Learning: Developing quantum advantage in ML tasks
  • Quantum Feature Maps: Designing expressive quantum embeddings
  • Barren Plateau Analysis: Understanding trainability in quantum circuits
  • Quantum Kernel Methods: Exploring quantum-enhanced kernel machines
  • NISQ Algorithms: Practical quantum computing applications

🛠️ Installation

pip install quantum-data-embedding-suite

For development:

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

📚 Documentation Structure

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for details on:

  • Submitting bug reports and feature requests
  • Contributing code and documentation
  • Adding new embedding types or metrics
  • Improving performance and compatibility

📄 Citation

If you use this package in your research, please cite:

@software{quantum_data_embedding_suite,
  title={Quantum Data Embedding Suite: Advanced Classical-to-Quantum Data Embedding for QML},
  author={Krishna Bajpai},
  year={2025},
  url={https://github.com/krish567366/quantum-data-embedding-suite}
}

📞 Support

Author: Krishna Bajpai
License: MIT
Version: 0.1.0