コード例 #1
0
def answer_q10():
    """
    generates clusters, figure out distoration and return x & y values
    """
    data_sources = [viz.DATA_111_URL, viz.DATA_290_URL, viz.DATA_896_URL]
    x_vals = range(6, 21)
    y_vals_hier = {}
    y_vals_kmean = {}
    for idx in range(len(data_sources)):
        # 0. Generate data_field & cluster_list
        clust_list, data_table = closest_pair.create_cluster_list(
            data_sources[idx])
        y_vals_hier[idx] = []
        # 1. calculate values for hierarchical - decreasing order
        for clust_size in reversed(x_vals):
            clust_list = closest_pair.hierarchical_clustering(clust_list,
                                                              clust_size)
            clust_error = closest_pair.compute_distortions(clust_list,
                                                           data_table)
            y_vals_hier[idx].insert(0, clust_error)
        # 2. calculate values for kmeans - decreasing order
        y_vals_kmean[idx] = []
        for clust_size in x_vals:
            clust_list, data_table = closest_pair.create_cluster_list(
                data_sources[idx])
            clust_list = closest_pair.kmeans_clustering(clust_list,
                                                        clust_size, 5)
            clust_error = closest_pair.compute_distortions(clust_list,
                                                           data_table)
            y_vals_kmean[idx].append(clust_error)
    return x_vals, y_vals_hier, y_vals_kmean
コード例 #2
0
def test_q7():
    clust_list, data_table = create_cluster_list(URL)
    print(closest_pair.compute_distortions(clust_list, data_table))
    test1 = closest_pair.hierarchical_clustering(clust_list, 16)
    print(closest_pair.compute_distortions(test1, data_table))
    clust_list, data_table = create_cluster_list(URL)
    test2 = closest_pair.kmeans_clustering(clust_list, 16, 5)
    print(closest_pair.compute_distortions(test2, data_table))
コード例 #3
0
"""
code to answer question 7
"""

import closest_pair
import alg_project3_viz

URL = alg_project3_viz.DATA_111_URL

clust_list, data_table = closest_pair.create_cluster_list(URL)
test1 = closest_pair.hierarchical_clustering(clust_list,9)
print(closest_pair.compute_distortions(test1, data_table))
clust_list, data_table = closest_pair.create_cluster_list(URL)
test2 = closest_pair.kmeans_clustering(clust_list,9,5)
print(closest_pair.compute_distortions(test2, data_table))