---

Genetic Algorithms, K-Means, and Particle Swarm Optimization

This project implements K-Means Clustering, Genetic Algorithms (GA), and Particle Swarm Optimization (PSO) in both sequential and parallel versions. The implementations are written in C++ and Haskell, leveraging parallelism with Charm++ for C++ and Haskell Strategies for Haskell. The goal is to provide a comparative framework for these optimization and clustering algorithms across different languages and parallelization techniques.

Project Overview

• Algorithms: K-Means, Genetic Algorithm, Particle Swarm Optimization
• Languages: C++ and Haskell
• Parallelism:
  • C++: Uses Charm++ for parallel execution.
  • Haskell: Uses Haskell Strategies for parallel execution.
• Structure: Each algorithm has its own directory, with subdirectories for the implementation technology (C++/Charm++ or Haskell) and further divided into parallel and sequential versions.

Directory Structure

.
├── kmeans
│   ├── charm++
│   │   ├── parallel    # Parallel K-Means in C++ with Charm++
│   │   │   └── Makefile
│   │   └── sequential  # Sequential K-Means in C++
│   │       └── Makefile
│   └── haskell
│       ├── parallel    # Parallel K-Means in Haskell
│       │   └── kmeans.cabal
│       └── sequential  # Sequential K-Means in Haskell
│           └── kmeans.cabal
├── ga
│   ├── charm++
│   │   ├── parallel    # Parallel GA in C++ with Charm++
│   │   │   └── Makefile
│   │   └── sequential  # Sequential GA in C++
│   │       └── Makefile
│   └── haskell
│       ├── parallel    # Parallel GA in Haskell
│       │   └── ga.cabal
│       └── sequential  # Sequential GA in Haskell
│           └── ga.cabal
├── pso
│   ├── charm++
│   │   ├── parallel    # Parallel PSO in C++ with Charm++
│   │   │   └── Makefile
│   │   └── sequential  # Sequential PSO in C++
│   │       └── Makefile
│   └── haskell
│       ├── parallel    # Parallel PSO in Haskell
│       │   └── pso.cabal
│       └── sequential  # Sequential PSO in Haskell
│           └── pso.cabal
└── README.md

Prerequisites

For C++ Implementations

• g++: C++ compiler (e.g., GCC).
• Charm++: Required for parallel C++ implementations. Install it by following the Charm++ installation guide.

For Haskell Implementations

• GHC: Glasgow Haskell Compiler (version compatible with base >=4.13).
• Cabal: Haskell package manager (version >= 1.10).
• Dependencies: Install required Haskell libraries:

  cabal update
  cabal install random split normaldistribution criterion-measurement deepseq parallel

Building and Running

C++ Implementations (with Makefile)

Each charm++/parallel and charm++/sequential directory contains a Makefile to compile the code.

1. Navigate to the desired directory:

   cd ga/charm++/parallel  # Example for parallel GA

2. Build the project:

   make

   This compiles the source files (e.g., main.cpp, ga.cpp) into an executable named after the algorithm (e.g., ga).

3. Run the executable:

   ./ga

4. Clean up (optional):

   make clean

   This removes object files and the executable.

Notes for Charm++ (Parallel C++)

• Ensure Charm++ is installed and configured.
• You may need to adjust the Makefile to include Charm++-specific flags (e.g., linking with charmc instead of g++).

Haskell Implementations (with Cabal)

Each haskell/parallel and haskell/sequential directory contains a .cabal file (e.g., ga.cabal) to build and run the code.

1. Navigate to the desired directory:

   cd ga/haskell/parallel  # Example for parallel GA

2. Build the project:

   cabal build

   This compiles the Haskell source files (e.g., Main.hs, GA.hs) with optimizations (-O2) and threading support.

3. Run the executable:

   cabal run ga

   For parallel execution, you can specify runtime options:

   cabal run ga -- +RTS -N -RTS

   -N enables all available CPU cores for parallelism.

4. Clean up (optional):

   cabal clean

Example Files

• Makefile (C++): Compiles .cpp files into object files and links them into an executable.
• ga.cabal (Haskell): Specifies dependencies, source directories, and GHC options like -threaded for parallelism.

Contributing

Feel free to fork this repository, submit issues, or contribute improvements via pull requests.

License

See the LICENSE file for details.

---