Project for Parallel Computing course. Sequential and parallel implementations of K-Means algorithm in C++ with OpenMP and CUDA. The sequential K-Means version is compared with two parallel versions, tests include execution times and the speedup of each version.
More information can be found in the report of the project: Report.
- Clone the repo.
git clone https://github.com/elia-mercatanti/parallel-k-means
- Build with CMake.
- For testing K-Means algorithm with all implementations to search clusters on a dataset, pass two arguments:
parallel_kmeans <dataset file path> <number of clusters>
- For generating random datasets according to global variables, pass no arguments:
parallel_kmeans <>
- Elia Mercatanti - GitHub: elia-mercatanti
- Marco Calamai - GitHub: marcocalamai
Licensed under the term of MIT License.