Parallel implementation of K-Mean algorithm with OpenMP.
Compile with:
g++ -fopenmp main.cpp -o project_kmean
Execute with:
./project_kmean <input.csv> <kmean_variant> <num_threads> <k> <initialization_alg>
<input.csv>
: input data file name (10 features).<kmean_variant>
: choose the clustering variant ('kmean' or 'kmedoids').<num_threads>
: number of threads.<k>
: number of clusters.<initialization_alg>
: set how to define the initial centroids ('random' or 'kmeanpp').
The algorithm returns the execution times of the different parts and the silhouette score.