def run_example():
"""
Load a data table, compute a list of clusters and
plot a list of clusters
Set DESKTOP = True/False to use either matplotlib or simplegui
"""
data_table = load_data_table(DATA_3108_URL)
singleton_list = []
for line in data_table:
singleton_list.append(alg_cluster.Cluster(set([line[0]]), line[1], line[2], line[3], line[4]))
#cluster_list = sequential_clustering(singleton_list, 15)
#print "Displaying", len(cluster_list), "sequential clusters"
cluster_list = HW3.hierarchical_clustering(singleton_list, 15)
print "Displaying", len(cluster_list), "hierarchical clusters"
# cluster_list = HW3.kmeans_clustering(singleton_list,9, 5)
# print "Displaying", len(cluster_list), "k-means clusters"
#draw the clusters using matplotlib or simplegui
if DESKTOP:
alg_clusters_matplotlib.plot_clusters(data_table, cluster_list, False)
def experiment_10(a):
concatenate={}
concatenate[111]=DATA_111_URL
concatenate[290]=DATA_290_URL
concatenate[896]=DATA_896_URL
data_table = load_data_table(concatenate[a])
##for the hierarchical
singleton_list = []
for line in data_table:
singleton_list.append(alg_cluster.Cluster(set([line[0]]), line[1], line[2], line[3], line[4]))
for ele in x:
singleton_list=HW3.hierarchical_clustering(singleton_list,ele)
distortion=0
for cluster in singleton_list:
distortion+=cluster.cluster_error(data_table)
if a==111:
y_111_H.append(distortion)
elif a==290:
y_290_H.append(distortion)
else:
y_896_H.append(distortion)
##K-means
singleton_list = []
for line in data_table:
singleton_list.append(alg_cluster.Cluster(set([line[0]]), line[1], line[2], line[3], line[4]))
for ele in x:
singleton_list=HW3.kmeans_clustering(singleton_list,ele,5)
distortion=0
for cluster in singleton_list:
distortion+=cluster.cluster_error(data_table)
if a==111:
y_111_K.append(distortion)
elif a==290:
y_290_K.append(distortion)
else:
y_896_K.append(distortion)
if a==111:
plt.plot(x,y_111_H)
plt.plot(x,y_111_K)
elif a==290:
plt.plot(x,y_290_H)
plt.plot(x,y_290_K)
else:
plt.plot(x,y_896_H)
plt.plot(x,y_896_K)
plt.xlabel('Number of the clusters ')
plt.ylabel('Total Distortion')
plt.legend(['Hierarchical Clustering','K-mean clustering'])
plt.title('Distortion comparison of two clustering method-'+str(a)+'counties')
plt.grid(True)
plt.savefig("HW3_"+str(a)+".png")
plt.show()
