Skip to main content
. 2025 Jul 1;25(13):4111. doi: 10.3390/s25134111
Algorithm 2 Our proposed data-size reduction approach.
Input: training_X, testing_X, testing_y, number_components, number_clusters
Output: training_X_centroids, testing_X_pca
  •   1:

    pca_var ← PCA(number_components)

  •   2:

    pca_var.fit(training_X)

  •   3:

    training_X_pca ← pca_var.transform(training_X)

  •   4:

    testing_X_pca ← pca_var.transform(testing_X)

  •   5:

    kmeans_var ← KMeans(number_clusters)

  •   6:

    kmeans_var.fit(training_X_pca)

  •   7:

    training_X_centroids ← kmeans_var.cluster_centers

  •   8:

    return training_X_centroids, testing_X_pca