CUDA Levenshtein-Damerau Distance Calculator

A high-performance C++ library for computing Levenshtein-Damerau distances between string pairs using both CPU and GPU implementations.

Features

• CPU Implementation: Optimized Levenshtein-Damerau distance algorithm with O(n*m) time complexity
• CUDA GPU Implementation: Parallel GPU kernel for batch processing of string pairs
• Comprehensive Testing: Full test suite covering all edit operations (insertion, deletion, substitution, transposition)
• Performance Benchmarks: Detailed performance comparison between CPU and GPU implementations
• CMake Build System: Cross-platform build support with CUDA integration

Building the Project

Prerequisites

• CMake 3.18 or higher
• CUDA 12.0 or higher
• C++20 compatible compiler (GCC 13.x or Clang 15+)
• NVIDIA GPU with compute capability 5.2 or higher

Compiler Compatibility Note

Build Instructions

mkdir build && cd build
cmake ..
make

License

This project is provided as-is for educational and research purposes.