예제 #1
0
def generate_clusters_multiple(subject_num, feature_list_1, feature_list_2, feature_list_3, n_clusters = 5):

  labels_1 = kmeans.perform_kMeans_clustering_analysis(feature_list_1, n_clusters)
  labels_2 = kmeans.perform_kMeans_clustering_analysis(feature_list_2, n_clusters)
  labels_3 = kmeans.perform_kMeans_clustering_analysis(feature_list_3, n_clusters)

  return labels_1, labels_2, labels_3
예제 #2
0
def generate_clusters(subject_num, feature_list_1, feature_list_2, feature_list_3):

  n_clusters = 5
  TR = project_config.TR

  labels_1 = kmeans.perform_kMeans_clustering_analysis(feature_list_1, n_clusters)
  labels_2 = kmeans.perform_kMeans_clustering_analysis(feature_list_2, n_clusters)
  labels_3 = kmeans.perform_kMeans_clustering_analysis(feature_list_3, n_clusters)

  labels_list = [labels_1, labels_2, labels_3]

  result_labels = kmeans.merge_n_clusters(labels_list, n_clusters, labels_1.shape)

  return result_labels
예제 #3
0
def generate_clusters_multiple(subject_num,
                               feature_list_1,
                               feature_list_2,
                               feature_list_3,
                               n_clusters=5):

    labels_1 = kmeans.perform_kMeans_clustering_analysis(
        feature_list_1, n_clusters)
    labels_2 = kmeans.perform_kMeans_clustering_analysis(
        feature_list_2, n_clusters)
    labels_3 = kmeans.perform_kMeans_clustering_analysis(
        feature_list_3, n_clusters)

    return labels_1, labels_2, labels_3
예제 #4
0
def generate_clusters(subject_num, feature_list_1, feature_list_2,
                      feature_list_3):

    n_clusters = 5
    TR = project_config.TR

    labels_1 = kmeans.perform_kMeans_clustering_analysis(
        feature_list_1, n_clusters)
    labels_2 = kmeans.perform_kMeans_clustering_analysis(
        feature_list_2, n_clusters)
    labels_3 = kmeans.perform_kMeans_clustering_analysis(
        feature_list_3, n_clusters)

    labels_list = [labels_1, labels_2, labels_3]

    result_labels = kmeans.merge_n_clusters(labels_list, n_clusters,
                                            labels_1.shape)

    return result_labels
def single_subject_kmeans(standard_source_prefix, subject_num, task_num):

  residuals, in_brain_mask = preprocessing_pipeline(subject_num, task_num, standard_source_prefix)

  labels = kmeans.perform_kMeans_clustering_analysis(residuals.reshape((-1, residuals.shape[-1])), 6)

  b_vols = np.zeros(in_brain_mask.shape)
  b_vols[in_brain_mask] = labels
  b_vols[~in_brain_mask] = np.nan

  return b_vols