def clusterize(self):
print("\nclusterize")
self.process_clustering_data()
c = Clustering(self.clustering_df)
c.k_means(2)
c.k_means(3)
c.k_means(4)
def main(fn, clusters_no):
geo_locs = []
#read location data from csv file and store each location as a Point(latit,longit) object
df = pd.read_csv(fn)
for index, row in df.iterrows():
loc_ = Point(float(row['LAT']), float(row['LON'])) #tuples for location
geo_locs.append(loc_)
#run k_means clustering
cluster = Clustering(geo_locs, clusters_no)
flag = cluster.k_means(False)
if flag == -1:
print("Error in arguments!")
else:
#clustering results is a list of lists where each list represents one cluster
print("Clustering results:")
cluster.print_clusters(cluster.clusters)
####################################################################
print('Minkowski Weighted PAM')
start = time.time()
[u, medoids, weights, ite, dist_tmp] = cl.mwpam(data, 3, 1.1, False, 10)
print('Time elapsed: ', time.time()-start)
print('Accuracy: ', cl.my_math.compare_categorical_vectors(u, y)[0])
####################################################################
print('Minkowski Weighted PAM (Initialized with Minkowski Build)')
start = time.time()
[u, medoids, weights, ite, dist_tmp] = cl.mwpam(data, 3, 1.1)
print('Time elapsed: ', time.time()-start)
print('Accuracy: ', cl.my_math.compare_categorical_vectors(u, y)[0])
####################################################################
print('K-Means')
start = time.time()
[u, centroids, ite, dist_tmp] = cl.k_means(data, 3, replicates=10)
print('Time elapsed: ', time.time()-start)
print('Accuracy: ', cl.my_math.compare_categorical_vectors(u, y)[0])
####################################################################
print('iK-Means')
start = time.time()
[u, centroids, ite, dist_tmp, init_centroids] = cl.ik_means(data, 3)
print('Time elapsed: ', time.time()-start)
print('Accuracy: ', cl.my_math.compare_categorical_vectors(u, y)[0])
####################################################################
print('WK-Means')
start = time.time()
[u, centroids, weights, ite, dist_tmp] = cl.wk_means(data, 3, 1.1, replicates=10)
print('Time elapsed: ', time.time()-start)
print('Accuracy: ', cl.my_math.compare_categorical_vectors(u, y)[0])
####################################################################
def subspace_cluster(datapath,
ypath,
algorithm,
k,
sep=",",
preprocess_method="standard",
replicates=10,
max_ite=100,
beta=2,
init_weights=None,
init_centroids=None,
init_weights_method="random",
is_sparse=0,
threshold=0.9,
l=2,
minDeviation=0.1,
A=30,
B=10):
np.seterr(all='raise')
cl = Clustering()
data_ori = np.genfromtxt(datapath, delimiter=sep)
y = np.genfromtxt(ypath)
data = data_preprocess(data_ori, preprocess_method)
if algorithm == 'K-Means':
print 'using K-Means'
start = time.time()
[u, centroids, ite, dist_tmp] = cl.k_means(data, k, replicates)
time_elapsed = time.time() - start
acc = cl.my_math.compare_categorical_vectors(u, y)[0]
print 'Time elapsed: ', time_elapsed
print 'Accuracy: ', acc
return [u, centroids, ite, dist_tmp], time_elapsed, acc
elif algorithm == 'iK-Means':
print 'using iK-Means'
start = time.time()
[u, centroids, ite, dist_tmp, init_centroids] = cl.ik_means(data, k)
time_elapsed = time.time() - start
acc = cl.my_math.compare_categorical_vectors(u, y)[0]
print 'Time elapsed: ', time_elapsed
print 'Accuracy: ', acc
return [u, centroids, ite, dist_tmp, init_centroids], time_elapsed, acc
elif algorithm == 'WK-Means':
print 'using WK-Means'
start = time.time()
[u, centroids, weights, ite,
dist_tmp] = cl.wk_means(data,
k,
beta,
init_centroids=init_centroids,
init_weights=init_weights,
replicates=replicates,
max_ite=max_ite,
init_weights_method=init_weights_method,
is_sparse=is_sparse,
threshold=threshold)
time_elapsed = time.time() - start
acc = cl.my_math.compare_categorical_vectors(u, y)[0]
print 'Time elapsed: ', time_elapsed
print 'Accuracy: ', acc
return [u, centroids, weights, ite, dist_tmp], time_elapsed, acc
elif algorithm == 'MWK-Means':
print 'using MWK-Means'
start = time.time()
[u, centroids, weights, ite,
dist_tmp] = cl.mwk_means(data,
k,
beta,
init_centroids=init_centroids,
init_weights=init_weights,
replicates=replicates,
max_ite=max_ite,
init_weights_method=init_weights_method,
is_sparse=is_sparse,
threshold=threshold)
time_elapsed = time.time() - start
acc = cl.my_math.compare_categorical_vectors(u, y)[0]
print 'Time elapsed: ', time_elapsed
print 'Accuracy: ', acc
return [u, centroids, weights, ite, dist_tmp], time_elapsed, acc
elif algorithm == 'iMWK-Means':
print 'using iMWK-Means'
start = time.time()
[u, centroids, weights, ite, dist_tmp] = cl.imwk_means(data, beta, k)
time_elapsed = time.time() - start
acc = cl.my_math.compare_categorical_vectors(u, y)[0]
print 'Time elapsed: ', time_elapsed
print 'Accuracy: ', acc
return [u, centroids, weights, ite, dist_tmp], time_elapsed, acc
elif algorithm == 'proclus':
print 'using proclus'
start = time.time()
M, D, A, acc = cl.proclus(data,
y,
k,
l,
minDeviation=minDeviation,
A=A,
B=B,
niters=max_ite)
time_elapsed = time.time() - start
print 'Time elapsed: ', time_elapsed
print "Accuracy: %.4f" % acc
return [M, D, A], time_elapsed, acc
