def test_gam_ranked_mushrooms_runtime(self):
local_attribution_path = 'data/mushrooms.csv'
g = GAM(attributions_path=local_attribution_path,
n_clusters=3,
cluster_method="ranked medoids")
g.generate()
print("Ranked Medoids Runtime: ", g.duration)
def test_gam_kernel_mushrooms_runtime(self):
local_attribution_path = '../data/mice_protein.csv'
g = GAM(attributions_path=local_attribution_path,
n_clusters=3,
cluster_method="kernel medoids",
dataset='mice_protein')
g.generate()
print("Kernel Runtime: ", g.duration)
def test_gam_bandit_mushrooms_runtime(self):
local_attribution_path = 'data/mushrooms.csv'
g = GAM(attributions_path=local_attribution_path,
n_clusters=3,
cluster_method="bandit pam",
dataset='mushrooms')
g.generate()
print("BanditPAM Runtime: ", g.duration)
def test_gam_bandit_mice_runtime_700_samples(self):
local_attribution_path = '../data/mice_protein.csv'
g = GAM(attributions_path=local_attribution_path,
n_clusters=3,
cluster_method="bandit pam",
num_samp=700,
dataset='mice_protein')
g.generate()
print("BanditPAM 700 Runtime: ", g.duration)
def test_gam_bandit_crime_runtime_1000_samples(self):
local_attribution_path = '../data/crime.csv'
g = GAM(attributions_path=local_attribution_path,
n_clusters=3,
cluster_method="bandit pam",
num_samp=1000,
dataset='crime')
g.generate()
print("BanditPAM 1000 Runtime: ", g.duration)
def test_gam_bandit_wine_runtime(self):
print(
"-- test_gam_bandit_wine_runtime -------------------------------------------------------"
)
gam = GAM()
local_attribution_path = 'data/wine.csv'
g = GAM(attributions_path=local_attribution_path,
n_clusters=3,
cluster_method="bandit pam")
g.generate()
print("BanditPAM Runtime: ", g.duration, "\n")
def test_gam_parallel_wine_bestk(self):
local_attribution_path = '../data/wine_clean.csv'
bestClusterNumber = 0
bestScore = -2
for k in range(2,5):
g = GAM(attributions_path=local_attribution_path, n_clusters=k, cluster_method="bandit pam", dataset='wine')
g.generate()
if g.avg_silhouette_score > bestScore:
bestScore = g.avg_silhouette_score
bestClusterNumber = k
print("Best Number of Clusters: ", bestClusterNumber)
print("Best Silhouette Score:, ", bestScore)
def test_gam_parallel_mice_bestk(self):
local_attribution_path = '../data/mice_protein.csv'
bestClusterNumber = 0
bestScore = -2
for k in range(2, 5):
g = GAM(attributions_path=local_attribution_path,
n_clusters=k,
cluster_method="parallel medoids",
dataset='mice_protein')
g.generate()
if g.avg_silhouette_score > bestScore:
bestScore = g.avg_silhouette_score
bestClusterNumber = k
print("Best Number of Clusters: ", bestClusterNumber)
print("Best Silhouette Score:, ", bestScore)
def test_gam_kmedoids_wine_runtime(self):
local_attribution_path = '../data/wine_clean.csv'
g = GAM(attributions_path=local_attribution_path, n_clusters=3, cluster_method=None)
g.generate()
print("Original Medoids Algorithm Runtime: ", g.duration)
def test_gam_bandit_wine_runtime(self):
local_attribution_path = '../data/wine_clean.csv'
g = GAM(attributions_path=local_attribution_path, n_clusters=3, cluster_method="bandit pam", dataset='wine')
g.generate()
print("BanditPAM Runtime: ", g.duration)
def test_gam_parallel_wine_runtime(self):
local_attribution_path = '../data/wine_clean.csv'
g = GAM(attributions_path=local_attribution_path, n_clusters=3, cluster_method="parallel medoids")
g.generate()
print("Parallel Medoids Runtime: ", g.duration)
def test_gam_bandit_mushrooms_runtime(self):
gam = GAM()
local_attribution_path = 'data/mushroom-attributions-200-samples.csv'
g = GAM(attributions_path=local_attribution_path, n_clusters=3, cluster_method="bandit pam")
g.generate()
print("BanditPAM Runtime: ", g.duration)