def test_complex(repeat, diff_ratio, num_trial, size):
"""
diff_ratio used for simulator
num_trial = num_trials to find center - use minimum pair
"""
sim_avg=0.0
num_successful_clustering=0
cluster_score_avg=0.0
inaccurate_min_pair_counter = 0
for i in range(repeat):
test_matrix,origin_tracker=haplotype_simulator(1000,size,100,3,diff_ratio)
# find_minimum_pair_new(test_matrix, size, num_trial)
test_center, sim_score, center_one_row, center_two_row = find_minimum_pair_new(test_matrix, size, num_trial)
if origin_tracker[center_one_row] == origin_tracker[center_two_row]:
inaccurate_min_pair_counter = inaccurate_min_pair_counter + 1
sim_avg = sim_avg + sim_score
print
ms_vector = learn.cluster.MeanShift(seeds=test_center).fit_predict(test_matrix)
cluster_score = cluster_checker(origin_tracker,ms_vector)
if cluster_score == 0.0:
num_successful_clustering = num_successful_clustering + 1
pair_ind=""
if origin_tracker[center_one_row] == origin_tracker[center_two_row]:
pair_ind = "WRONG PAIR"
else:
pair_ind = "RIGHT PAIR"
print "Trial # ", i, "Error Percentage ",cluster_score, "n(NONZERO) ", np.count_nonzero(test_center[0]), np.count_nonzero(test_center[1]), pair_ind
cluster_score_avg = cluster_score_avg + cluster_score
print "average simliarty = ", sim_avg/repeat, "and # successful clustering = ", num_successful_clustering
print "average success rate = ", 1-(cluster_score_avg/repeat), "# wrong min pair = ", inaccurate_min_pair_counter
def test(repeat):
sim_avg=0.0;
num_successful_clustering=0;
for i in range(repeat):
test_matrix,origin_tracker=haplotype_simulator(1000,1000,100,2,3)
print origin_tracker
test_center=[]
test_center.append(test_matrix[0])
test_center.append(test_matrix[500])
sim_avg = sim_avg + similarity(test_center)
# test_bandwidth=learn.cluster.estimate_bandwidth(test_matrix, quantile=0.5)
ms_vector = learn.cluster.MeanShift(seeds=test_center).fit_predict(test_matrix)
if ms_vector[0] == ms_vector[251] and ms_vector[523]==ms_vector[809]:
num_successful_clustering=num_successful_clustering+1
print "average simliarty = ", sim_avg/repeat, "and # successful clustering = ", num_successful_clustering
def test(repeat):
sim_avg = 0.0
num_successful_clustering = 0
for i in range(repeat):
test_matrix, origin_tracker = haplotype_simulator(
1000, 1000, 100, 2, 3)
print origin_tracker
test_center = []
test_center.append(test_matrix[0])
test_center.append(test_matrix[500])
sim_avg = sim_avg + similarity(test_center)
# test_bandwidth=learn.cluster.estimate_bandwidth(test_matrix, quantile=0.5)
ms_vector = learn.cluster.MeanShift(
seeds=test_center).fit_predict(test_matrix)
if ms_vector[0] == ms_vector[251] and ms_vector[523] == ms_vector[809]:
num_successful_clustering = num_successful_clustering + 1
print "average simliarty = ", sim_avg / repeat, "and # successful clustering = ", num_successful_clustering
def test_complex(repeat, diff_ratio, num_trial, size):
"""
diff_ratio used for simulator
num_trial = num_trials to find center - use minimum pair
"""
sim_avg = 0.0
num_successful_clustering = 0
cluster_score_avg = 0.0
inaccurate_min_pair_counter = 0
for i in range(repeat):
test_matrix, origin_tracker = haplotype_simulator(
1000, size, 100, 3, diff_ratio)
# find_minimum_pair_new(test_matrix, size, num_trial)
test_center, sim_score, center_one_row, center_two_row = find_minimum_pair_new(
test_matrix, size, num_trial)
if origin_tracker[center_one_row] == origin_tracker[center_two_row]:
inaccurate_min_pair_counter = inaccurate_min_pair_counter + 1
sim_avg = sim_avg + sim_score
print
ms_vector = learn.cluster.MeanShift(
seeds=test_center).fit_predict(test_matrix)
cluster_score = cluster_checker(origin_tracker, ms_vector)
if cluster_score == 0.0:
num_successful_clustering = num_successful_clustering + 1
pair_ind = ""
if origin_tracker[center_one_row] == origin_tracker[center_two_row]:
pair_ind = "WRONG PAIR"
else:
pair_ind = "RIGHT PAIR"
print "Trial # ", i, "Error Percentage ", cluster_score, "n(NONZERO) ", np.count_nonzero(
test_center[0]), np.count_nonzero(test_center[1]), pair_ind
cluster_score_avg = cluster_score_avg + cluster_score
print "average simliarty = ", sim_avg / repeat, "and # successful clustering = ", num_successful_clustering
print "average success rate = ", 1 - (
cluster_score_avg /
repeat), "# wrong min pair = ", inaccurate_min_pair_counter
