for dt in rrule(MONTHLY, dtstart=start_date, until=end_date):
target_timestamp = np.append(target_timestamp, dt.strftime("%Y-%m"))
print('target_timestamp', target_timestamp)
x=input()
total_borrow_data = pd.DataFrame()
total_return_data = pd.DataFrame()
for count in range(int(len(target_timestamp) / 12)):
title_name = ""
total_borrow_data.drop(total_borrow_data.index, inplace=True)
total_return_data.drop(total_return_data.index, inplace=True)
for x, y in zip(borrow_file_name[12 * count: 12 * (count + 1)], return_file_name[12 * count: 12 * (count + 1)]):
borrow_data = LoadData.load_month_borrow_data(x)
return_data = LoadData.load_month_return_data(y)
total_borrow_data = pd.concat((total_borrow_data, borrow_data), axis=0)
total_return_data = pd.concat((total_return_data, return_data), axis=0)
# print('borrow_data\n', total_borrow_data.shape)
# print('return_data\n', total_return_data.shape)
np_data = [np.array(total_borrow_data).T, np.array(total_return_data).T]
# 進行 K-medoids 分群並視覺化
# 執行兩次，一次borrow 一次return
condition = ['Borrow', 'Return']
for data_set, choose in zip(np_data, condition):
k_number = [x for x in range(3, 11, 1)]
evaluation_score = []
cluster_record = {}
for k in k_number:
