C++ API Reference for Intel® Data Analytics Acceleration Library 2020 Update 1
Base classes parameters for computing initial centroids for K-Means algorithm. More...
Parameter | ( | size_t | _nClusters, |
size_t | _offset = 0 , |
||
size_t | _seed = 777777 |
||
) |
Parameter constructor
[in] | _nClusters | Number of clusters |
[in] | _offset | Offset in the total data set specifying the start of a block stored on a given local node |
[in] | _seed | Seed for generating random numbers for the initialization |
Constructs parameters of the algorithm that computes initial centroids for K-Means algorithm by copying another parameters object
[in] | other | Parameters of K-Means algorithm |
engines::EnginePtr engine |
Engine to be used for generating random numbers for the initialization
size_t nClusters |
Number of clusters
size_t nRounds |
Kmeans|| only. Number of rounds for k-means||. (oversamplingFactor*nRounds) > 1 is a requirement.
size_t nRowsTotal |
Total number of rows in the data set
size_t offset |
Offset in the total data set specifying the start of a block stored on a given local node
double oversamplingFactor |
Kmeans|| only. A fraction of nClusters being chosen in each of nRounds of kmeans||.\ L = nClusters* oversamplingFactor points are sampled in a round.
size_t seed |
Seed for generating random numbers for the initialization
For more complete information about compiler optimizations, see our Optimization Notice.