def _initialise(self, filename, i): # Starts by reading the TC data (tables_x.txt), create a population of parameter sets
for file in filename:
self.i = i
self.tables.append(read_data.main(file,i))
i += 1
text1 = "gender then lastname ascending\n--------------------------\n"
text2 = "\ndateofbirth ascending\n--------------------------\n"
text3 = "\nlastname descending\n--------------------------\n"
if not os.path.exists(
os.path.dirname(PROJECT_ABSOLUTE_PATH + "\\output\\output.txt")):
try:
os.makedirs(
os.path.dirname(PROJECT_ABSOLUTE_PATH + "\\output\\output.txt"))
except OSError as exception:
if exception.errno != errno.EEXIST:
raise
"""
Execution of first sorting
"""
final_list1 = read_data.main()
GenderLastname = gender_lastname_sort.gender_lastname_sort(final_list1)
"""
Execution of second sorting
"""
final_list2 = read_data.main()
for i in final_list2:
i[3] = datetime.strptime(i[3], '%m/%d/%Y')
Datebirth = datebirth_sort.datebirth_sort(final_list2)
for i in Datebirth:
i[3] = datetime.strftime(i[3], '%m/%d/%Y')
"""
Execution of third sorting
"""
final_list3 = read_data.main()
Lastname = lastname_sort.lastname_sort(final_list3)
def main():
path = 'data/Sam_full_ROM_set_1_method_3.xlsx'
data = read_data.main(path)
ax = read_data.group_plot(data)
current_value = np.zeros([4])
pre_1st_value = np.zeros([4])
pre_2nd_value = np.zeros([4])
pre_3rd_value = np.zeros([4])
current_diff = np.zeros([4])
pre_1st_diff = np.zeros([4])
pre_2nd_diff = np.zeros([4])
pre_3rd_diff = np.zeros([4])
cycle_boo = 0
cycle_count = 0
new_peak = np.zeros([4])
new_high = np.zeros([4])
current_peak = np.zeros([4])
old_peak = np.zeros([4])
current_start_value = np.zeros([4])
prev_start_value = np.zeros([4])
end_value = np.zeros([4])
sign = np.zeros([4])
end_sign = np.zeros([4])
start_direction = np.zeros([4])
current_sign = np.zeros([4])
prev_sign = np.zeros([4])
start_sign = np.zeros([4])
mag = np.zeros([4])
peak_peak = np.zeros([4])
peak_boo = np.zeros([4])
current_loc = np.zeros([4])
high_loc = np.zeros([4])
loc = np.zeros([4])
current_marked_peak = np.zeros([4])
current_max_start_diff = 0
current_max_start_index = 0
end_time = 0
interval = 20
for i_index, value in enumerate(data):
current_value = value[0:4]
current_loc = np.repeat(value[4], 4)
current_time = value[4]
current_diff = current_value - pre_1st_value
current_sign = current_diff / abs(current_diff)
peak_boo = np.zeros([4])
sum_diff = 0
for sum in enumerate(current_diff):
sum_diff += abs(sum[1])
check_count = 0
if min(abs(prev_start_value)) != 0 and min(abs(end_sign)) != 0:
for k_index in enumerate(current_value):
temp_index = k_index[0]
if current_sign[temp_index] == end_sign[temp_index] * (-1):
check_count += 1
else:
check_count = 4
if cycle_boo == 0 and current_time - end_time > 500:
if check_count >= 3:
for j_index in enumerate(current_diff):
index = j_index[0]
if abs(current_max_start_diff) < abs(current_diff[index]) \
and abs(current_diff[index]) > 1 \
and pre_1st_diff[index] != 0:
current_max_start_diff = abs(current_diff[index])
current_max_start_index = index
for j_index in enumerate(current_value):
index = j_index[0]
# Peak finding
if current_sign[index] == prev_sign[index]:
new_high[index] = current_value[index]
high_loc[index] = current_loc[index]
elif current_sign[index] != prev_sign[index] and cycle_boo > 0:
if current_sign[index] == -1:
sign[index] = 1
else:
sign[index] = -1
if sign[index] * (current_value[index] - new_high[index]) > 0:
new_peak[index] = current_value[index]
loc[index] = current_loc[index]
peak_boo[index] = 1
else:
new_peak[index] = new_high[index]
loc[index] = high_loc[index]
peak_boo[index] = 1
if new_peak[index] > current_peak[index]:
current_peak[index] = new_peak[index]
mag[index] = current_peak[index] - current_start_value[index]
if abs(mag[index]) > 30 \
and abs(current_peak[index] - current_marked_peak[index]) > 30 \
and current_sign[index] != prev_sign[index]:
peak_boo[index] = 1
current_marked_peak[index] = current_peak[index]
if cycle_boo == 1:
cycle_boo = 2
ax[index].axvline(x=loc[index], color='#45F112')
end_value[index] = current_peak[index] - current_value[index]
if index == current_max_start_index:
if current_max_start_diff > 2 \
and cycle_boo == 0 \
and 0 < abs(pre_1st_diff[index]) < abs(current_diff[index]):
check_count = 0
if prev_start_value[index] != 0 and end_sign[index] != 0:
for k_index in enumerate(current_value):
temp_index = k_index[0]
if current_sign[
temp_index] == end_sign[temp_index] * (-1):
check_count += 1
else:
check_count = 4
if check_count >= 3:
if prev_start_value[index] != 0 and end_sign[
index] != 0:
cycle_boo = 1
for k_index in enumerate(current_value):
temp_index = k_index[0]
start_sign[temp_index] = current_sign[
temp_index]
current_start_value[
temp_index] = current_value[temp_index]
ax[temp_index].axvline(
x=current_loc[temp_index],
color='#d62728') # Red
prev_start_value[
temp_index] = current_start_value[
temp_index]
else:
cycle_boo = 1
for k_index in enumerate(current_value):
temp_index = k_index[0]
start_sign[temp_index] = current_sign[
temp_index]
current_start_value[
temp_index] = current_value[temp_index]
prev_start_value[
temp_index] = current_start_value[
temp_index]
ax[temp_index].axvline(x=current_loc[index],
color='#d62728') # Red
if cycle_boo == 2:
check_count = 0
for k_index in enumerate(end_value):
temp_index = k_index[0]
if abs(end_value[temp_index]) >= 0.85 * abs(mag[temp_index]) \
and 0 < abs(pre_1st_diff[temp_index]) < 15 \
and 0 < abs(current_diff[temp_index]) < abs(pre_1st_diff[temp_index]) \
and current_sign[temp_index] == start_sign[temp_index] * (-1):
check_count += 1
if check_count >= 3:
cycle_boo = 0
cycle_count += 1
current_peak = np.zeros([4])
current_marked_peak = np.zeros([4])
current_max_start_diff = 0
for k_index in enumerate(current_value):
temp_index = k_index[0]
end_sign[temp_index] = current_sign[temp_index]
ax[temp_index].axvline(x=current_loc[temp_index],
color='#d67e27')
end_time = value[4]
prev_sign = current_sign
pre_3rd_value = pre_2nd_value
pre_2nd_value = pre_1st_value
pre_1st_value = current_value
pre_3rd_diff = pre_2nd_diff
pre_2nd_diff = pre_1st_diff
pre_1st_diff = current_diff
plt.tight_layout()
plt.show()
print(cycle_count)
pass
import pandas as pd
import numpy as np
import math
import operator
import matplotlib.pyplot as plt
import sys
import read_data
read_data.main()
theta = None
features = None
column_names = [
'CRIM', 'ZN', 'INDUS', 'CHAS', 'NOX', 'RM', 'AGE', 'DIS', 'RAD', 'TAX',
'PTRATIO', 'B', 'LSTAT', 'MEDV'
]
def mean(array):
return float(sum(array) / len(array))
def stdev(numbers):
meanVal = mean(numbers)
variance = sum([pow(x - meanVal, 2)
for x in numbers]) / float(len(numbers) - 1)
return float(math.sqrt(variance))
def variance(numbers):
avg = mean(numbers)
variance = sum([pow(x - avg, 2)
def main():
path = 'data/Sam_full_ROM_set_1_method_1.xlsx'
data = read_data.main(path)
ax = read_data.group_plot(data)
current_value = np.zeros([4])
pre_1st_value = np.zeros([4])
pre_2nd_value = np.zeros([4])
pre_3rd_value = np.zeros([4])
current_diff = np.zeros([4])
pre_1st_diff = np.zeros([4])
pre_2nd_diff = np.zeros([4])
pre_3rd_diff = np.zeros([4])
cycle_boo = np.zeros([4])
cycle_count = np.zeros([4])
new_peak = np.zeros([4])
new_high = np.zeros([4])
current_peak = np.zeros([4])
old_peak = np.zeros([4])
current_start_value = np.zeros([4])
prev_start_value = np.zeros([4])
end_value = np.zeros([4])
sign = np.zeros([4])
end_sign = np.zeros([4])
start_direction = np.zeros([4])
current_sign = np.zeros([4])
prev_sign = np.zeros([4])
start_sign = np.zeros([4])
mag = np.zeros([4])
peak_peak = np.zeros([4])
peak_boo = np.zeros([4])
current_loc = np.zeros([4])
high_loc = np.zeros([4])
loc = np.zeros([4])
current_marked_peak = np.zeros([4])
interval = 20
for i_index, value in enumerate(data):
current_value = value[0:4]
current_loc = np.repeat(value[4], 4)
current_diff = current_value - pre_1st_value
current_sign = current_diff / abs(current_diff)
peak_boo = np.zeros([4])
sum_diff = 0
for sum in enumerate(current_diff):
sum_diff += abs(sum[1])
for j_index in enumerate(current_value):
index = j_index[0]
if current_sign[index] == prev_sign[index]:
new_high[index] = current_value[index]
high_loc[index] = current_loc[index]
elif current_sign[index] != prev_sign[index] and cycle_boo[
index] > 0:
if current_sign[index] == -1:
sign[index] = 1
else:
sign[index] = -1
if sign[index] * (current_value[index] - new_high[index]) > 0:
new_peak[index] = current_value[index]
loc[index] = current_loc[index]
peak_boo[index] = 1
else:
new_peak[index] = new_high[index]
loc[index] = high_loc[index]
peak_boo[index] = 1
if new_peak[index] > current_peak[index]:
current_peak[index] = new_peak[index]
mag[index] = current_peak[index] - current_start_value[index]
if abs(mag[index]) > 30 \
and abs(current_peak[index] - current_marked_peak[index]) > 20 \
and current_sign[index] != prev_sign[index]:
peak_boo[index] = 1
current_marked_peak[index] = current_peak[index]
if cycle_boo[index] == 1:
cycle_boo[index] = 2
ax[index].axvline(x=loc[index], color='#45F112')
end_value[index] = current_peak[index] - current_value[index]
if abs(current_diff[index]) > 5:
if cycle_boo[index] == 0 \
and 0 < abs(pre_1st_diff[index]) < abs(current_diff[index]):
start_sign[index] = current_sign[index]
if prev_start_value[index] != 0 and end_sign[index] != 0:
if current_sign[index] == end_sign[index] * (-1):
current_start_value[index] = current_value[index]
cycle_boo[index] = 1
ax[index].axvline(x=current_loc[index],
color='#d62728') #Red
prev_start_value[index] = current_start_value[
index]
else:
current_start_value[index] = current_value[index]
cycle_boo[index] = 1
ax[index].axvline(x=current_loc[index],
color='#d62728') # Red
prev_start_value[index] = current_start_value[index]
if cycle_boo[index] == 2 \
and 0 < abs(current_diff[index]) < 2 \
and 0 < abs(current_diff[index]) < abs(pre_1st_diff[index]) \
and abs(end_value[index]) >= 0.9 * abs(mag[index]) \
and current_sign[index] == start_sign[index] * (-1):
cycle_boo[index] = 0
cycle_count[index] += 1
end_sign[index] = current_sign[index]
print(
[index, current_peak[index] - current_start_value[index]])
current_peak[index] = 0
current_marked_peak[index] = 0
ax[index].axvline(x=current_loc[index], color='#d67e27')
prev_sign = current_sign
pre_3rd_value = pre_2nd_value
pre_2nd_value = pre_1st_value
pre_1st_value = current_value
pre_3rd_diff = pre_2nd_diff
pre_2nd_diff = pre_1st_diff
pre_1st_diff = current_diff
plt.tight_layout()
plt.show()
print(cycle_count)
pass