From SW's laptop

Error Analysis.py

@@ -0,0 +1,291 @@
+import pandas as pd
+import numpy as np
+import matplotlib.pyplot as plt
+
+
+'''
+데이터 구조: Error_overall.csv
+'File', 'Data_usage',
+'RMSE_hyper_original', 'RMSE_hyper_nonlinear',
+'Final_error_hyper_original', 'Final_error_hyper_nonlinear']
+'''
+
+# CSV 파일 읽기
+df_overall = pd.read_csv('Error_overall.csv', encoding='euc-kr')
+
+# 통계량 저장소
+# 열 mean / median / percentile
+# 행 RMSE (O) / RMSE (NL) / FE (O) / FE (NL)
+statistic =[]
+
+count = 0
+
+# 최종 성토 단계에서 각 침하 데이터 사용 영역에 대해서 다음을 수행
+for data_usage in range(20, 100, 10):
+
+    # 전체 Error 분석을 위한 Dataframe 설정
+    df_overall_sel = df_overall.loc[df_overall['Data_usage'] == data_usage]
+
+    # RMSE 및 FE를 불러서 메모리에 저장
+    RMSE_hyper_original = df_overall_sel['RMSE_hyper_original'].to_numpy()
+    RMSE_hyper_nonlinear = df_overall_sel['RMSE_hyper_nonlinear'].to_numpy()
+    FE_hyper_original = df_overall_sel['Final_error_hyper_original'].to_numpy()
+    FE_hyper_nonlinear = df_overall_sel['Final_error_hyper_nonlinear'].to_numpy()
+
+    # 중앙값, 평균, 90% percentile 산정 및 저장
+    statistic.append([np.mean(RMSE_hyper_original),
+                      np.median(RMSE_hyper_original),
+                      np.percentile(RMSE_hyper_original, 90)])
+    statistic.append([np.mean(RMSE_hyper_nonlinear),
+                      np.median(RMSE_hyper_nonlinear),
+                      np.percentile(RMSE_hyper_nonlinear, 90)])
+    statistic.append([np.mean(FE_hyper_original),
+                      np.median(FE_hyper_original),
+                      np.percentile(FE_hyper_original, 90)])
+    statistic.append([np.mean(FE_hyper_nonlinear),
+                      np.median(FE_hyper_nonlinear),
+                      np.percentile(FE_hyper_nonlinear, 90)])
+
+    # 그래프 설정 (2 by 2)
+    fig, ((ax1, ax2), (ax3, ax4)) = plt.subplots(2, 2, figsize = (8, 8))
+
+    # 그래프 제목 설정
+    fig.suptitle('Histograms: ' + str(data_usage) +
+                 '% of Settlement Data Used in the Final Step')
+
+    # 각 Subplot의 제목 설정
+    ax1.set_xlabel('RMSE (Original Hyperbolic) (cm)')
+    ax2.set_xlabel('RMSE (Nonlinear Hyperbolic) (cm)')
+    ax3.set_xlabel('FE (Original Hyperbolic) (cm)')
+    ax4.set_xlabel('FE (Nonliner Hyperbolic) (cm)')
+
+    # 각 subplot에 히스토그램 작성
+    ax1.hist(RMSE_hyper_original, 5, density=True, facecolor='r', edgecolor='k', alpha=0.75)
+    ax2.hist(RMSE_hyper_nonlinear, 5, density=True, facecolor='b', edgecolor='k', alpha=0.75)
+    ax3.hist(FE_hyper_original, 5, density=True, facecolor='r', edgecolor='k', alpha=0.75)
+    ax4.hist(FE_hyper_nonlinear, 5, density=True, facecolor='b', edgecolor='k', alpha=0.75)
+
+    # 각 subplot을 포함한 리스트 설정
+    axes = [ax1, ax2, ax3, ax4]
+
+    # 공통 사항 적용
+    for i in range(len(axes)):
+        ax = axes[i]
+        ax.text(10, 0.4, 'Mean = ' + "{:0.2f}".format(statistic[count * 4 + i][0]) + '\n' +
+                'Median = ' + "{:0.2f}".format(statistic[count * 4 + i][1]) + '\n' +
+                '90% Percentile = ' + "{:0.2f}".format(statistic[count * 4 + i][2]))
+
+        ax.set_ylabel("Probability")
+        ax.grid(color="gray", alpha=.5, linestyle='--')
+        ax.tick_params(direction='in')
+        ax.set_xlim(0, 50)
+        ax.set_ylim(0, 0.5)
+
+    count = count + 1
+
+    # 그래프 저장 (SVG 및 PNG)
+    plt.savefig('error_analysis/error_nonstep(%i percent).png' % data_usage,
+                bbox_inches='tight')
+
+data_usages = range(20, 100, 10)
+statistic = np.array(statistic)
+
+fig, ((ax1, ax2), (ax3, ax4), (ax5, ax6)) = plt.subplots(3, 2, figsize = (8, 12))
+
+fig.suptitle("Original Hyperbolic vs. Nonlinear Hyperbolic")
+
+# mean rmse
+ax1.set_ylabel('Mean(RMSE) (cm)')
+ax1.plot(data_usages, statistic[0::4, 0], label = 'Original Hyperbolic')
+ax1.plot(data_usages, statistic[1::4, 0], label = 'Nonlinear Hyperbolic')
+ax1.legend()
+
+# mean fe
+ax2.set_ylabel('Mean(FE) (cm)')
+ax2.plot(data_usages, statistic[2::4, 0])
+ax2.plot(data_usages, statistic[3::4, 0])
+
+# median rmse
+ax3.set_ylabel('Median(RMSE) (cm)')
+ax3.plot(data_usages, statistic[0::4, 1])
+ax3.plot(data_usages, statistic[1::4, 1])
+
+# median fe
+ax4.set_ylabel('Median(FE) (cm)')
+ax4.plot(data_usages, statistic[2::4, 1])
+ax4.plot(data_usages, statistic[3::4, 1])
+
+# percentile rmse
+ax5.set_ylabel('90% Percentile(RMSE) (cm)')
+ax5.plot(data_usages, statistic[0::4, 2])
+ax5.plot(data_usages, statistic[1::4, 2])
+
+# percentile fe
+ax6.set_ylabel('90% Percentile(FE) (cm)')
+ax6.plot(data_usages, statistic[2::4, 2])
+ax6.plot(data_usages, statistic[3::4, 2])
+
+axes = [ax1, ax2, ax3, ax4, ax5, ax6]
+
+# 공통 사항 적용
+for ax in axes:
+    ax.set_xlabel("Data Usage (%)")
+    ax.grid(color="gray", alpha=.5, linestyle='--')
+    ax.tick_params(direction='in')
+    ax.set_xlim(0, 100)
+    ax.set_ylim(0, 50)
+
+# 그래프 저장 (SVG 및 PNG)
+plt.savefig('error_analysis/error_overall.png', bbox_inches='tight')
+
+
+'''
+데이터 구조: Error_multi_step.csv
+'File', 'Data_usage',
+'RMSE_hyper_original', 'RMSE_hyper_nonlinear', 'RMSE_step',
+'Final_error_hyper_original', 'Final_error_hyper_nonlinear', 'Final_error_step'
+'''
+
+df_multi_step = pd.read_csv('Error_multi_step.csv', encoding='euc-kr')
+
+# 통계량 저장소
+# 열 mean / median / percentile
+# 행 RMSE (O) / RMSE (NL) / RMSE(S) / FE (O) / FE (NL) / FE (S)
+statistic2 =[]
+count = 0
+
+# 최종 성토 단계에서 각 침하 데이터 사용 영역에 대해서 다음을 수행
+for data_usage in range(20, 100, 10):
+
+    # 전체 Error 분석을 위한 Dataframe 설정
+    df_multi_step_sel = df_multi_step.loc[df_multi_step['Data_usage'] == data_usage]
+
+    # RMSE 및 FE를 불러서 메모리에 저장
+    RMSE_hyper_original = df_multi_step_sel['RMSE_hyper_original'].to_numpy()
+    RMSE_hyper_nonlinear = df_multi_step_sel['RMSE_hyper_nonlinear'].to_numpy()
+    RMSE_step = df_multi_step_sel['RMSE_step'].to_numpy()
+    FE_hyper_original = df_multi_step_sel['Final_error_hyper_original'].to_numpy()
+    FE_hyper_nonlinear = df_multi_step_sel['Final_error_hyper_nonlinear'].to_numpy()
+    FE_step = df_multi_step_sel['Final_error_step'].to_numpy()
+
+    # 중앙값, 평균, 90% percentile 산정 및 저장
+    statistic2.append([np.mean(RMSE_hyper_original),
+                       np.median(RMSE_hyper_original),
+                       np.percentile(RMSE_hyper_original, 90)])
+    statistic2.append([np.mean(RMSE_hyper_nonlinear),
+                       np.median(RMSE_hyper_nonlinear),
+                       np.percentile(RMSE_hyper_nonlinear, 90)])
+    statistic2.append([np.mean(RMSE_step),
+                       np.median(RMSE_step),
+                       np.percentile(RMSE_step, 90)])
+    statistic2.append([np.mean(FE_hyper_original),
+                       np.median(FE_hyper_original),
+                       np.percentile(FE_hyper_original, 90)])
+    statistic2.append([np.mean(FE_hyper_nonlinear),
+                       np.median(FE_hyper_nonlinear),
+                       np.percentile(FE_hyper_nonlinear, 90)])
+    statistic2.append([np.mean(FE_step),
+                       np.median(FE_step),
+                       np.percentile(FE_step, 90)])
+
+    # 그래프 설정 (2 by 2)
+    fig, ((ax1, ax2, ax3), (ax4, ax5, ax6)) = plt.subplots(2, 3, figsize = (12, 8))
+
+    # 그래프 제목 설정
+    fig.suptitle('Histograms: ' + str(data_usage) +
+                 '% of Settlement Data Used in the Final Step')
+
+    # 각 Subplot의 제목 설정
+    ax1.set_xlabel('RMSE (Original Hyperbolic) (cm)')
+    ax2.set_xlabel('RMSE (Nonlinear Hyperbolic) (cm)')
+    ax3.set_xlabel('RMSE (Step) (cm)')
+    ax4.set_xlabel('FE (Original Hyperbolic) (cm)')
+    ax5.set_xlabel('FE (Nonliner Hyperbolic) (cm)')
+    ax6.set_xlabel('FE (Step) (cm)')
+
+    # 각 subplot에 히스토그램 작성
+    ax1.hist(RMSE_hyper_original, 5, density=True, facecolor='r', edgecolor='k', alpha=0.75)
+    ax2.hist(RMSE_hyper_nonlinear, 5, density=True, facecolor='b', edgecolor='k', alpha=0.75)
+    ax3.hist(RMSE_step, 5, density=True, facecolor='g', edgecolor='k', alpha=0.75)
+    ax4.hist(FE_hyper_original, 5, density=True, facecolor='r', edgecolor='k', alpha=0.75)
+    ax5.hist(FE_hyper_nonlinear, 5, density=True, facecolor='b', edgecolor='k', alpha=0.75)
+    ax6.hist(FE_step, 5, density=True, facecolor='g', edgecolor='k', alpha=0.75)
+
+    # 각 subplot을 포함한 리스트 설정
+    axes = [ax1, ax2, ax3, ax4, ax5, ax6]
+
+    # 공통 사항 적용
+    for i in range(len(axes)):
+        ax = axes[i]
+        ax.text(10, 0.4, 'Mean = ' + "{:0.2f}".format(statistic2[count * 6 + i][0]) + '\n' +
+                'Median = ' + "{:0.2f}".format(statistic2[count * 6 + i][1]) + '\n' +
+                '90% Percentile = ' + "{:0.2f}".format(statistic2[count * 6 + i][2]))
+        ax.set_ylabel("Probability")
+        ax.grid(color="gray", alpha=.5, linestyle='--')
+        ax.tick_params(direction='in')
+        ax.set_xlim(0, 50)
+        ax.set_ylim(0, 0.5)
+
+    count = count + 1
+
+    # 그래프 저장 (SVG 및 PNG)
+    plt.savefig('error_analysis/error_step(%i percent).png' % data_usage,
+                bbox_inches='tight')
+
+
+data_usages = range(20, 100, 10)
+statistic2 = np.array(statistic2)
+
+fig, ((ax1, ax2), (ax3, ax4), (ax5, ax6)) = plt.subplots(3, 2, figsize = (8, 12))
+
+fig.suptitle("Hyperbolic vs. Step loading")
+
+# mean rmse
+ax1.set_ylabel('Mean(RMSE) (cm)')
+ax1.plot(data_usages, statistic2[0::6, 0], label = 'Original Hyperbolic')
+ax1.plot(data_usages, statistic2[1::6, 0], label = 'Nonlinear Hyperbolic')
+ax1.plot(data_usages, statistic2[2::6, 0], color='k', label = 'Step Loading')
+ax1.legend()
+
+# mean fe
+ax2.set_ylabel('Mean(FE) (cm)')
+ax2.plot(data_usages, statistic2[3::6, 0])
+ax2.plot(data_usages, statistic2[4::6, 0])
+ax2.plot(data_usages, statistic2[5::6, 0], color='k')
+
+# median rmse
+ax3.set_ylabel('Median(RMSE) (cm)')
+ax3.plot(data_usages, statistic2[0::6, 1])
+ax3.plot(data_usages, statistic2[1::6, 1])
+ax3.plot(data_usages, statistic2[2::6, 1], color='k')
+
+# median fe
+ax4.set_ylabel('Median(FE) (cm)')
+ax4.plot(data_usages, statistic2[3::6, 1])
+ax4.plot(data_usages, statistic2[4::6, 1])
+ax4.plot(data_usages, statistic2[5::6, 1], color='k')
+
+# percentile rmse
+ax5.set_ylabel('90% Percentile(RMSE) (cm)')
+ax5.plot(data_usages, statistic2[0::6, 2])
+ax5.plot(data_usages, statistic2[1::6, 2])
+ax5.plot(data_usages, statistic2[2::6, 2], color='k')
+
+# percentile fe
+ax6.set_ylabel('90% Percentile(FE) (cm)')
+ax6.plot(data_usages, statistic2[3::6, 2])
+ax6.plot(data_usages, statistic2[4::6, 2])
+ax6.plot(data_usages, statistic2[5::6, 2], color='k')
+
+axes = [ax1, ax2, ax3, ax4, ax5, ax6]
+
+# 공통 사항 적용
+for ax in axes:
+    ax.set_xlabel("Data Usage (%)")
+    ax.grid(color="gray", alpha=.5, linestyle='--')
+    ax.tick_params(direction='in')
+    ax.set_xlim(0, 100)
+    ax.set_ylim(0, 50)
+
+# 그래프 저장 (SVG 및 PNG)
+plt.savefig('error_analysis/error_step.png', bbox_inches='tight')

batch_process.py

@@ -2,11 +2,17 @@ import settle_prediction_steps_main
 import pandas as pd
 import os
 
-input_dir = 'data'
-output_dir = 'output'
+input_dir = 'data_1'
+output_dir = 'output_1'
 input_files = []
 
-df = pd.DataFrame(columns=['File', 'Data_usage',
+df_overall = pd.DataFrame(columns=['File', 'Data_usage',
+                           'RMSE_hyper_original',
+                           'RMSE_hyper_nonlinear',
+                           'Final_error_hyper_original',
+                           'Final_error_hyper_nonlinear'])
+
+df_multi_step = pd.DataFrame(columns=['File', 'Data_usage',
                            'RMSE_hyper_original',
                            'RMSE_hyper_nonlinear',
                            'RMSE_step',
@@ -14,16 +20,33 @@ df = pd.DataFrame(columns=['File', 'Data_usage',
                            'Final_error_hyper_nonlinear',
                            'Final_error_step'])
 
+
 for (root, directories, files) in os.walk(input_dir):
     for file in files:
         file_path = os.path.join(root, file)
         input_files.append(file_path)
 
 for input_file in input_files:
-    for i in range(20, 100, 20):
-        ERROR = settle_prediction_steps_main.run_settle_prediction(input_file,
-                                                              output_dir, i, 100, False, False)
-        df.loc[len(df.index)] = [input_file, i, ERROR[0], ERROR[1], ERROR[2],
-                                 ERROR[3], ERROR[4], ERROR[5]]
+    for i in range(20, 100, 10):
 
-df.to_csv('Error.csv')
+        RETURN_VALUES = settle_prediction_steps_main.\
+            run_settle_prediction(input_file, output_dir, i, 100, False, False)
+
+        df_overall.loc[len(df_overall.index)] = [input_file, i,
+                                                 RETURN_VALUES[0],
+                                                 RETURN_VALUES[1],
+                                                 RETURN_VALUES[3],
+                                                 RETURN_VALUES[4]]
+
+        if RETURN_VALUES[6]:
+            df_multi_step.loc[len(df_overall.index)] = [input_file, i,
+                                                        RETURN_VALUES[0],
+                                                        RETURN_VALUES[1],
+                                                        RETURN_VALUES[2],
+                                                        RETURN_VALUES[3],
+                                                        RETURN_VALUES[4],
+                                                        RETURN_VALUES[5]]
+
+# 에러 파일 출력
+df_overall.to_csv('Error_overall.csv')
+df_multi_step.to_csv('Error_multi_step.csv')

data/1_S-12.csv

@@ -1,144 +0,0 @@
-Time,Settle,Surcharge
-0,0,3.52
-2,1.2,3.52
-5,3.1,3.52
-8,5.4,3.52
-15,9.6,3.52
-21,21.5,4.835
-26,28.7,4.835
-28,31.4,4.835
-34,38.3,4.835
-37,41.7,4.835
-40,44.6,4.835
-43,47.3,4.835
-47,51.2,4.835
-50,54.1,4.835
-54,57.3,4.835
-57,59.7,4.835
-61,63,4.835
-64,64.7,4.835
-68,66.9,4.835
-70,68.2,4.835
-72,69.6,4.835
-75,71.5,4.835
-76,72.2,4.835
-77,72.9,4.835
-78,73.5,4.835
-79,74.1,4.835
-82,75.3,4.835
-83,75.7,4.835
-84,76.1,4.835
-85,76.5,4.835
-89,78.4,4.835
-90,78.9,4.835
-92,79.9,4.835
-96,81.8,4.835
-98,82.8,4.835
-99,83.2,4.835
-105,85.8,4.835
-106,86.2,4.835
-107,86.6,4.835
-110,87.8,4.835
-112,88.6,4.835
-114,89.4,4.835
-117,90.6,4.835
-119,91.4,4.835
-121,92.2,4.835
-128,95.2,4.835
-131,96.4,4.835
-134,97.6,4.835
-138,99,4.835
-140,99.6,4.835
-147,101.4,4.835
-149,102,4.835
-153,103.2,4.835
-155,103.8,4.835
-159,105.1,4.835
-162,106,4.835
-166,110.4,6.05
-167,111.3,6.05
-168,112.2,6.05
-169,113.1,6.05
-170,114,6.05
-173,116.7,6.05
-174,117.5,6.05
-177,119.9,6.05
-181,123.1,6.05
-182,123.9,6.05
-183,124.7,6.05
-184,125.4,6.05
-187,127.5,6.05
-188,128.1,6.05
-189,128.7,6.05
-190,129.2,6.05
-191,129.7,6.05
-194,131.2,6.05
-195,131.7,6.05
-196,132.2,6.05
-197,132.7,6.05
-198,133.1,6.05
-201,134.3,6.05
-202,134.7,6.05
-203,135.1,6.05
-204,135.5,6.05
-205,135.9,6.05
-208,137.1,6.05
-211,138.3,6.05
-215,139.5,6.05
-217,140.1,6.05
-219,140.7,6.05
-222,141.6,6.05
-225,142.5,6.05
-229,143.7,6.05
-233,144.9,6.05
-237,146.1,6.05
-239,146.7,6.05
-243,147.9,6.05
-246,148.8,6.05
-250,150,6.05
-253,150.9,6.05
-257,152.1,6.05
-261,153.3,6.05
-264,154.2,6.05
-267,155.1,6.05
-271,155.9,6.05
-275,156.7,6.05
-278,157.3,6.05
-281,157.9,6.05
-285,158.7,6.05
-292,160.1,6.05
-295,160.7,6.05
-299,161.5,6.05
-302,162.1,6.05
-306,162.9,6.05
-309,163.5,6.05
-313,164.3,6.05
-316,164.9,6.05
-320,165.7,6.05
-324,166.1,6.05
-327,166.4,6.05
-329,166.6,6.05
-334,167.1,6.05
-337,167.4,6.05
-341,167.8,6.05
-344,168.1,6.05
-348,168.5,6.05
-351,168.8,6.05
-355,169.2,6.05
-358,169.5,6.05
-362,169.9,6.05
-369,170.6,6.05
-372,170.9,6.05
-376,171.3,6.05
-380,171.7,6.05
-383,172,6.05
-386,172.3,6.05
-390,172.7,6.05
-393,173,6.05
-397,173.4,6.05
-400,173.7,6.05
-404,174,6.05
-407,174.2,6.05
-411,174.5,6.05
-414,174.7,6.05
-422,175.2,6.05

data/1_SP-11.csv

@@ -1,125 +0,0 @@
-Time,Settle,Surcharge
-0,0,1.5
-5,17.4,1.5
-7,23.9,1.5
-11,32.2,1.5
-14,41.7,1.5
-21,64.1,1.5
-28,72.5,1.5
-35,78.8,1.5
-42,93.3,1.5
-48,102.5,1.5
-53,108,3.002
-54,109.2,3.002
-55,110.4,3.002
-56,111.6,3.002
-59,117.3,3.002
-60,119.2,3.002
-61,121.1,3.002
-62,122.7,3.002
-67,130.2,3.002
-68,131.9,3.002
-69,133.6,3.002
-70,135.4,3.002
-74,141.4,3.002
-75,142.9,3.002
-76,144.4,3.002
-77,146.2,3.002
-80,149.2,3.002
-81,150.2,3.002
-82,151.2,3.002
-83,152.2,3.002
-91,162.8,3.002
-98,170,3.002
-105,177,3.002
-112,182.4,3.002
-115,185,3.002
-117,186.5,3.002
-118,187.3,3.002
-122,202.9,4.095
-124,210.5,4.095
-125,214.5,4.095
-126,218.6,4.095
-129,222.4,4.095
-130,223.7,4.095
-131,225,4.095
-132,226.3,4.095
-133,227.5,4.095
-136,231.7,4.095
-137,233.1,4.095
-138,234.5,4.095
-139,235.9,4.095
-140,237.3,4.095
-143,240.7,4.095
-147,245.5,4.095
-151,249.7,4.095
-154,252.8,4.095
-158,257.8,4.095
-161,261.1,4.095
-164,264.1,4.095
-168,268,4.095
-172,272.2,4.095
-175,275.5,4.095
-181,283.5,4.095
-192,293.5,4.095
-195,296.2,4.095
-199,301.3,4.095
-202,304.6,4.095
-209,311.1,4.095
-216,316,4.095
-223,322.3,4.095
-230,326.5,4.095
-237,331.6,4.095
-244,336.5,4.095
-251,341.2,4.095
-258,346.1,4.095
-266,350.9,4.095
-273,354,4.095
-280,356,4.095
-286,358,4.095
-294,360.9,4.095
-300,363,5.256
-301,363.4,5.256
-304,365.8,5.256
-305,366.5,5.256
-306,367.2,5.256
-307,367.9,5.256
-308,368.5,5.256
-311,369.5,5.256
-312,369.8,5.256
-313,370.1,5.256
-314,370.4,5.256
-327,377.4,5.256
-329,378.5,5.256
-336,381.8,5.256
-343,385.5,5.256
-350,388.4,5.256
-357,391.1,5.256
-364,394.1,5.256
-371,397.1,5.256
-377,399.5,5.256
-385,401.4,5.256
-388,402.3,5.256
-389,402.6,5.256
-390,402.9,5.256
-391,403.2,5.256
-392,403.5,5.256
-395,404.4,5.256
-397,405,5.256
-398,405.3,5.256
-402,406.5,5.256
-404,407.1,5.256
-405,407.4,5.256
-406,407.6,5.256
-409,408.2,5.256
-411,408.6,5.256
-419,410.2,5.256
-420,410.5,5.256
-425,411.5,5.256
-426,411.7,5.256
-434,413.3,5.256
-440,414.5,5.256
-447,415.9,5.256
-455,417.5,5.256
-461,418.7,5.256
-468,420,5.256

data/1_SP-23.csv

@@ -1,125 +0,0 @@
-Time,Settle,Surcharge
-0,0,1.5
-5,6.7,1.5
-8,10.2,1.5
-14,20.5,1.5
-22,28.2,1.5
-29,43.2,1.5
-35,47.4,1.5
-43,57.1,1.5
-50,63.5,1.5
-57,70.5,1.5
-64,77.7,1.5
-70,83.5,1.5
-78,92.4,1.5
-85,97.6,1.5
-92,103.5,1.5
-99,108.4,1.5
-106,113.49,1.5
-113,118.4,1.5
-116,118.4,2.871
-120,122.8,2.871
-124,123,2.871
-125,125.2,2.871
-126,127.4,2.871
-127,129.6,2.871
-130,132.9,2.871
-131,134,2.871
-132,135.1,2.871
-133,136.2,2.871
-144,145.6,2.871
-145,146.5,2.871
-146,147.4,2.871
-147,148.3,2.871
-154,153.4,2.871
-161,159.2,2.871
-168,163.9,2.871
-175,169.8,2.871
-182,174.9,2.871
-189,180.7,2.871
-196,185.4,2.871
-203,191.2,2.871
-210,196.1,3.606
-218,201.7,3.606
-225,205.9,3.606
-232,210.1,3.606
-238,213.7,3.606
-246,217.9,3.606
-253,221.2,3.606
-260,225.2,3.606
-266,228.4,3.606
-281,239.2,3.606
-288,243.1,3.606
-295,246.6,3.606
-302,250.4,3.606
-309,254,3.606
-316,257.3,3.606
-323,260.5,3.606
-329,263.3,3.606
-337,266.4,3.606
-344,269,3.606
-350,271.2,3.606
-358,274,3.606
-361,275.2,3.606
-363,276,3.606
-371,279.2,3.606
-372,279.6,3.606
-377,281.1,3.606
-378,281.4,3.606
-383,282.9,3.606
-384,283.2,3.606
-385,283.5,3.606
-386,283.8,3.606
-389,284.7,3.606
-390,285,3.606
-391,285.4,3.606
-392,285.8,3.606
-397,287.4,3.606
-398,287.7,3.606
-399,288,3.606
-403,289.2,3.606
-404,289.5,3.606
-405,289.8,3.606
-406,290.1,3.606
-407,290.4,3.606
-410,291,3.606
-411,291.2,3.606
-413,291.6,3.606
-420,293.1,3.606
-421,293.4,3.606
-424,294.3,3.606
-431,296.4,3.606
-435,298.4,4.896
-438,302,4.896
-441,305.1,4.896
-445,309.4,4.896
-448,313.7,4.896
-455,320.1,4.896
-462,325.4,4.896
-469,330.9,4.896
-473,334.3,4.896
-475,335.8,4.896
-476,336.5,4.896
-477,337.2,4.896
-480,338.8,4.896
-481,339.3,4.896
-483,340.3,4.896
-488,343.6,4.896
-497,348.6,4.896
-502,351.6,4.896
-504,352.3,4.896
-509,354.8,4.896
-512,356.1,4.896
-516,358.7,4.896
-518,359,4.896
-525,362.6,4.896
-530,364.6,4.896
-532,365.7,4.896
-540,368.4,4.896
-544,369.4,4.896
-547,370.2,4.896
-550,371,4.896
-553,372.4,4.896
-560,374.1,4.896
-565,375.2,4.896
-568,375.5,4.896

data/3_SP3-65.csv

@@ -1,183 +0,0 @@
-Time,Settle,Surcharge
-0,0,1.33
-4,7.7,1.33
-8,10.7,1.33
-11,20.4,1.33
-15,29.5,1.33
-17,36.5,1.33
-22,42.9,1.33
-24,48,1.33
-29,53.4,1.33
-31,58.6,1.33
-36,66,1.33
-38,69.5,1.33
-42,70.9,1.33
-45,77.7,1.33
-50,83.7,1.33
-53,89.2,1.33
-56,92.5,1.33
-59,95.4,1.33
-63,96.6,1.33
-66,97.4,1.33
-70,98.7,1.33
-73,100.6,1.33
-77,109.0583333,1.33
-80,115.825,1.33
-84,120.9,1.33
-87,122.6,1.33
-90,125.7,1.33
-94,129.7,1.33
-97,132.5,1.33
-100,135,1.33
-104,138,1.33
-107,141.5,1.33
-120,154.4,1.33
-122,156.2,1.33
-125,159.3,1.33
-128,162.2,1.33
-132,164.4,1.33
-135,167.3,1.33
-139,170.6,1.33
-142,173.3090909,1.33
-147,175.5,1.33
-150,177.6,1.33
-154,181.4,1.33
-156,183.8,1.33
-160,185.6,1.33
-163,188,1.33
-167,190.7,1.33
-170,192.6,1.33
-175,193.7,1.33
-178,195.2,1.33
-181,199.9,1.33
-184,201.9,1.33
-188,204,1.33
-191,205.9,1.33
-195,208.3,1.33
-198,210.3,1.33
-203,212.6,1.33
-207,214.3,1.33
-213,216.7,1.33
-216,218.5,1.33
-221,220.6,1.33
-224,222.3,1.33
-227,224.6,1.33
-230,226.4,1.33
-234,228.1,1.33
-238,230.4,1.33
-241,232.3,1.33
-245,233.9,1.33
-248,235,1.33
-252,236.9,1.33
-255,237.7,1.33
-258,239.2,1.33
-261,240.5,1.33
-265,242.2,1.33
-268,243.2,1.33
-272,244.4,1.33
-275,245.6,1.33
-279,247.1,1.33
-282,248,1.33
-286,249.3,1.33
-289,250.3,1.33
-293,252.1,1.33
-296,253.5,1.33
-300,255.4,1.33
-303,256.9,1.33
-308,259.2,1.33
-311,262.9,1.33
-315,268.3,1.33
-318,272.9,1.33
-322,279.5,1.33
-325,284.6,1.33
-328,288.7,1.33
-331,290.7,1.33
-338,292.9,1.33
-342,294.6,2.287
-345,295.9,2.287
-349,297.7,2.287
-352,299.2,2.287
-357,306.5,3.92
-361,311.9,3.92
-364,317,3.92
-367,321.9,3.92
-371,326.4,3.92
-374,329.9,3.92
-378,333.6,3.92
-381,336,3.92
-384,337.6,3.92
-387,338.9,3.92
-391,343.3,3.92
-394,346.1,3.92
-398,349.6,3.92
-401,352.7,3.92
-405,355.6,3.92
-408,357.5,3.92
-413,360.8,3.92
-415,362.1,3.92
-420,366.1,3.92
-423,368.3,3.92
-426,370.7,3.92
-429,372.7,3.92
-433,375.4,3.92
-436,377.4,3.92
-440,380,3.92
-444,382.3,3.92
-447,383.8,3.92
-450,385.5,3.92
-455,388.2,3.92
-457,389.1,3.92
-461,391.5,3.92
-464,393.2,3.92
-471,397.3,3.92
-475,399.8,3.92
-479,402.1,3.92
-483,404.2,3.92
-486,405.6,3.92
-489,407.3,3.92
-492,408.3,5.2
-496,412.3,5.2
-499,415.7,5.2
-504,418.7,5.2
-507,421,5.2
-511,423.7,5.2
-514,426,5.2
-517,428.6,5.2
-520,430.8,5.2
-525,433.7,5.2
-528,436.1,5.2
-532,437,5.2
-534,438.3,5.2
-538,440.8,5.2
-541,442,5.2
-545,444.1,5.2
-548,445.6,5.2
-552,448.1,5.2
-555,449.7,5.2
-559,451.9,5.2
-562,454.7,5.2
-566,456.5,5.2
-569,458.4,5.2
-573,460.2,5.2
-576,461.5,5.2
-580,463.2,5.2
-583,464.3,5.2
-588,468.2,5.2
-590,470.3,5.2
-595,470.7,5.2
-598,470.9,5.2
-602,471.9,5.2
-605,472.6,5.2
-609,473.8,5.2
-612,474.6,5.2
-615,475.2,5.2
-618,475.8,5.2
-623,476.8,5.2
-626,477.6,5.2
-629,478.3,5.2
-632,479.2,5.2
-636,480.6,5.2
-639,480.6,5.2
-643,480.7,5.2
-646,480.7,5.2
-650,480.7,5.2

data/3_SP3-68.csv

@@ -1,128 +0,0 @@
-Time,Settle,Surcharge
-0,0,1.887
-3,41,1.887
-6,61.5,1.887
-10,73.1,1.887
-13,81.6,1.887
-17,86.9,1.887
-20,91.9,1.887
-23,96.7,1.887
-26,101.6,1.887
-30,107.7,2.94
-33,113,2.94
-37,119.4,2.94
-40,122.8,2.94
-44,128.4,2.94
-47,131.9,2.94
-52,139.5,2.94
-54,140.5,2.94
-59,148,2.94
-62,151,2.94
-65,153.8,2.94
-68,156.6,2.94
-72,160,2.94
-75,162.5,2.94
-79,166,2.94
-83,169.1,2.94
-86,172,2.94
-89,174.7,2.94
-94,180.3,2.94
-96,182.1,2.94
-100,185.5,2.94
-103,188.4,2.94
-110,195.8,2.94
-114,198.7,2.94
-118,201.1,2.94
-122,205,2.94
-125,206.9,2.94
-128,209.3,2.94
-131,211.8,2.94
-135,214.2,2.94
-138,216.2,2.94
-143,217.4,2.94
-146,218.3,2.94
-150,219.1,2.94
-153,219.9,2.94
-156,220.9,2.94
-159,222,2.94
-164,227.7,2.94
-167,230.7,2.94
-171,234.8,3.48
-173,236.5,3.48
-177,239.7,3.48
-180,241.8,3.48
-184,243.9,3.48
-187,246.3,3.48
-191,249.6,3.48
-194,252.1,3.48
-198,255.5,3.48
-201,258.9,3.48
-205,261.2,3.48
-208,263.4,3.48
-212,265.8,3.48
-215,267.1,3.48
-219,268.9,3.48
-222,270.2,3.48
-227,273.4,3.48
-229,275,3.48
-234,277.3,3.48
-237,278.1,3.48
-241,280,3.48
-244,281.2,3.48
-248,283.2,3.48
-251,284.6,3.48
-254,286.1,3.48
-257,287.5,3.48
-262,289.5,3.48
-265,290.7,3.48
-268,292,3.48
-271,293.2,3.48
-275,294.7,3.48
-278,295.3,3.48
-282,296.3,3.48
-285,296.7,3.48
-289,297.6,3.48
-292,298.5,3.48
-296,299.8,3.48
-299,300.6,3.48
-303,302.2,3.48
-306,303.5,3.48
-310,312.8,5.608
-313,318.1,5.608
-317,322.8,6.794
-320,328.9,6.794
-324,335.3,6.794
-328,340.5,6.794
-331,342.9,6.794
-334,345.5,6.794
-339,348,6.794
-342,350.6,6.794
-345,353.1,6.794
-348,355.6,6.794
-352,358.9,6.794
-355,361,6.794
-362,365.6,6.794
-367,369,6.794
-370,370.5,6.794
-373,372.2,6.794
-376,373.8,6.794
-381,376.4,6.794
-384,378.5,6.794
-387,380.4,6.794
-390,382.4,6.794
-394,385.8,6.794
-397,388,6.794
-402,389.9,6.794
-405,390.9,6.794
-409,392.6,6.794
-412,394,6.794
-415,396.7,6.794
-418,397.5,6.794
-422,398.7,6.794
-425,399.1,6.794
-429,399.8,6.794
-432,400.9,6.794
-436,403.6,6.794
-439,405.6,6.794
-443,408.2,6.794
-446,409.4,6.794

data/4_S-11.csv

@@ -1,158 +0,0 @@
-Time,Settle,Surcharge
-0,0,3
-9,14.4,3
-14,31.4,3
-24,33.3,3
-27,34.1,3
-28,34.5,3
-29,34.7,3
-31,34.9,3
-34,35.3,3
-36,35.7,3
-38,36,3.5
-41,36.3,3.5
-43,45.8,3.5
-45,50.6,3.5
-48,55.3,3.5
-50,79.1,3.5
-52,91,3.5
-55,96.9,3.5
-57,99.9,3.5
-59,102.9,3.5
-66,134,3.5
-72,142,3.5
-78,150.7,3.5
-83,154.2,3.5
-84,155.9,3.5
-86,157.7,3.5
-90,160,3.5
-92,161.1,3.5
-94,162.3,3.5
-97,160.3,3.5
-98,159.3,3.5
-99,158.8,3.5
-101,158.3,3.5
-104,158.6,3.5
-106,158.6,3.5
-108,158.6,3.5
-111,158.7,3.5
-113,158.9,3.5
-115,160.2,3.5
-118,160.8,3.5
-120,161.5,3.5
-121,162.8,3.5
-122,164.2,3.5
-123,165.5,3.5
-125,166.9,3.5
-127,169.7,3.5
-127,176.6,4
-128,180.1,4
-129,181.9,4
-132,182.7,4
-134,183.6,4
-135,199.5,5.11
-136,207.5,5.11
-136,211.5,5.11
-139,213.5,5.11
-142,215.5,5.11
-143,219.8,5.11
-146,221.9,5.11
-148,224,5.11
-150,226.9,5.11
-153,228.4,5.11
-155,229.8,5.11
-157,236.7,5.11
-160,243.1,5.11
-161,249.1,5.11
-162,257.3,5.11
-163,261.6,5.11
-164,263.7,5.11
-167,264.7,5.11
-169,265.8,5.11
-171,272.2,5.11
-174,275.4,5.11
-176,278.6,5.11
-177,281.8,5.11
-178,283.4,5.11
-181,284.2,5.11
-182,285,5.11
-183,288.3,5.11
-184,289.9,5.11
-185,290.7,5.11
-188,291.1,5.11
-190,291.3,5.11
-192,291.5,5.11
-195,294.7,5.11
-197,297.9,5.11
-202,304.3,5.11
-210,309.3,5.11
-217,314.4,5.11
-224,321,5.11
-231,327.8,5.11
-246,336.9,5.11
-253,341.1,5.11
-259,344.7,5.11
-267,350.4,5.11
-273,354.5,5.11
-280,359,5.11
-287,363.3,5.11
-293,366.9,5.11
-301,371.1,5.11
-308,374.7,5.11
-318,379.7,5.11
-325,382.3,5.11
-332,384.8,5.11
-339,387.3,5.11
-346,389.8,5.11
-352,392.1,5.11
-356,395.3,5.11
-364,401.5,5.11
-373,408.3,5.11
-380,410.8,5.11
-386,413.5,5.11
-392,416.1,5.11
-399,419,5.11
-407,420.3,5.11
-412,421,5.11
-420,422,5.11
-428,423,5.11
-435,428.8,5.11
-440,431.4,5.11
-442,432.2,5.11
-444,433.7,5.61
-447,435,5.61
-448,435.3,5.61
-457,437.8,5.61
-464,439.6,5.61
-470,440.8,5.61
-476,441.8,5.61
-484,443.2,5.61
-491,444.3,5.61
-498,448.8,5.61
-505,453.3,5.61
-513,457.3,5.61
-520,459.5,5.61
-524,460.6,5.61
-531,461.9,5.61
-538,463.1,5.61
-546,464.4,5.61
-552,465.4,5.61
-560,466.7,5.61
-569,468,5.61
-576,469.1,5.61
-581,469.8,5.61
-591,471.3,5.61
-598,472.4,5.61
-605,473.5,5.61
-612,474.5,5.61
-619,475.3,5.61
-625,476.1,5.61
-631,476.9,5.61
-640,478,5.61
-644,478.4,5.61
-651,479,5.61
-658,479.6,5.61
-664,480.1,5.61
-672,480.7,5.61
-680,481.2,5.61
-689,481.7,5.61

settle_prediction_steps_main.py

@@ -10,8 +10,6 @@ The methodologies used are 1) superposition of time-settlement curves
 and 2) nonlinear regression for hyperbolic curves.
 """
 
-
-
 # =================
 # Import 섹션
 # =================
@@ -23,7 +21,6 @@ import matplotlib.pyplot as plt
 from scipy.optimize import least_squares
 
 
-
 # =================
 # Function 섹션
 # =================
@@ -32,14 +29,17 @@ from scipy.optimize import least_squares
 def generate_data_hyper(px, pt):
     return pt / (px[0] * pt + px[1])
 
+
 # 회귀식과 측정치와의 잔차 반환 (비선형 쌍곡선)
 def fun_hyper_nonlinear(px, pt, py):
     return pt / (px[0] * pt + px[1]) - py
 
+
 # 회귀식과 측정치와의 잔차 반환 (기존 쌍곡선)
 def fun_hyper_original(px, pt, py):
     return px[0] * pt + px[1] - pt / py
 
+
 # RMSE 산정
 def fun_rmse(py1, py2):
     mse = np.square(np.subtract(py1, py2)).mean()
@@ -48,8 +48,11 @@ def fun_rmse(py1, py2):
 
 def run_settle_prediction(input_file, output_dir,
                           final_step_predict_percent, additional_predict_percent,
-                          plot_show, print_values):
-
+                          plot_show,
+                          print_values,
+                          run_original_hyperbolic='True',
+                          run_nonlinear_hyperbolic='True',
+                          run_step_prediction='True'):
     # ====================
     # 파일 읽기, 데이터 설정
     # ====================
@@ -67,12 +70,9 @@ def run_settle_prediction(input_file, output_dir,
 
     # 데이터 닫기
 
-
     # 마지막 계측 데이터 index + 1 파악
     final_index = time.size
 
-
-
     # =================
     # 성토 단계 구분
     # =================
@@ -94,28 +94,23 @@ def run_settle_prediction(input_file, output_dir,
 
         # 만일 성토고의 변화가 있을 경우,
         if surcharge[index] != current_surcharge:
-
             step_end_index.append(index)
             step_start_index.append(index)
             current_surcharge = surcharge[index]
 
-
     # 마지막 성토 단계 끝 index 추가
     step_end_index.append(len(surcharge) - 1)
 
-
-
     # =================
     # 성토 단계 조정
     # =================
-    # 다음 경우를 제외하고 해석을 수행할 필요가 있음
-    # 성토고 유지 기간이 매우 짧을 경우
-    # 성토고 유지 기간 중 계측 데이터 수가 작을 경우
+    # 성토고 유지 기간이 매우 짧을 경우, 해석 단계에서 제외
 
-    #
+    # 조정 성토 시작 및 끝 인덱스 리스트 초기화
     step_start_index_adjust = []
     step_end_index_adjust = []
 
+    # 각 성토 단계 별로 분석
     for i in range(0, len(step_start_index)):
 
         # 현 단계 성토 시작일 / 끝일 파악
@@ -126,20 +121,22 @@ def run_settle_prediction(input_file, output_dir,
         step_span = step_end_date - step_start_date
         step_data_num = step_end_index[i] - step_start_index[i] + 1
 
-        if (step_span > 50 and step_data_num > 15):
+        # 성토고 유지일 및 데이터 개수 기준 적용
+        if step_span > 30 and step_data_num > 5:
             step_start_index_adjust.append((step_start_index[i]))
             step_end_index_adjust.append((step_end_index[i]))
 
+    #  성토 시작 및 끝 인덱스 리스트 업데이트
     step_start_index = step_start_index_adjust
     step_end_index = step_end_index_adjust
 
+    # 침하 예측을 수행할 단계 설정 (현재 끝에서 2단계 이용)
+    step_start_index = step_start_index[-2:]
+    step_end_index = step_end_index[-2:]
 
     # 성토 단계 횟수 파악 및 저장
     num_steps = len(step_start_index)
 
-
-
-
     # ===========================
     # 최종 단계 데이터 사용 범위 조정
     # ===========================
@@ -165,8 +162,6 @@ def run_settle_prediction(input_file, output_dir,
     final_step_monitor_end_index = step_end_index[num_steps - 1]
     step_end_index[num_steps - 1] = final_step_predict_end_index
 
-
-
     # =================
     # 추가 예측 구간 반영
     # =================
@@ -189,8 +184,6 @@ def run_settle_prediction(input_file, output_dir,
     # 마지막 인덱스값 재조정
     final_index = time.size
 
-
-
     # =============================
     # Settlement Prediction (Step)
     # =============================
@@ -234,7 +227,6 @@ def run_settle_prediction(input_file, output_dir,
         # 회귀분석 시행
         res_lsq_hyper_nonlinear \
             = least_squares(fun_hyper_nonlinear, x0,
-                            bounds=((0, 0),(np.inf, np.inf)),
                             args=(tm_this_step, sm_this_step))
 
         # 쌍곡선 계수 저장 및 출력
@@ -249,18 +241,16 @@ def run_settle_prediction(input_file, output_dir,
         sp_step[step_start_index[i]:final_index] = \
             sp_step[step_start_index[i]:final_index] + sp_to_end_update + s0_this_step
 
-
-
     # =========================================================
     # Settlement prediction (nonliner and original hyperbolic)
     # =========================================================
 
     # 성토 마지막 데이터 추출
-    tm_hyper = time[step_start_index[num_steps-1]:step_end_index[num_steps-1]]
-    sm_hyper = settle[step_start_index[num_steps-1]:step_end_index[num_steps-1]]
+    tm_hyper = time[step_start_index[num_steps - 1]:step_end_index[num_steps - 1]]
+    sm_hyper = settle[step_start_index[num_steps - 1]:step_end_index[num_steps - 1]]
 
     # 현재 단계 시작 부터 끝까지 시간 데이터 추출
-    time_hyper = time[step_start_index[num_steps-1]:final_index]
+    time_hyper = time[step_start_index[num_steps - 1]:final_index]
 
     # 초기 시점 및 침하량 산정
     t0_hyper = tm_hyper[0]
@@ -300,8 +290,6 @@ def run_settle_prediction(input_file, output_dir,
     sp_hyper_original = sp_hyper_original + s0_hyper
     time_hyper = time_hyper + t0_hyper
 
-
-
     # ==========
     # 에러 산정
     # ==========
@@ -327,23 +315,20 @@ def run_settle_prediction(input_file, output_dir,
 
     # RMSE 출력 (단계, 비선형 쌍곡선, 기존 쌍곡선)
     if print_values:
-        print("RMSE (Nonlinear Hyper + Step): %0.3f" %RMSE_step)
-        print("RMSE (Nonlinear Hyperbolic): %0.3f" %RMSE_hyper_nonlinear)
-        print("RMSE (Original Hyperbolic): %0.3f" %RMSE_hyper_original)
+        print("RMSE (Nonlinear Hyper + Step): %0.3f" % RMSE_step)
+        print("RMSE (Nonlinear Hyperbolic): %0.3f" % RMSE_hyper_nonlinear)
+        print("RMSE (Original Hyperbolic): %0.3f" % RMSE_hyper_original)
 
     # (최종 계측 침하량 - 예측 침하량) 계산
-
-    final_error_step = settle[-1] - sp_step_rmse[-1]
-    final_error_hyper_nonlinear = settle[-1] - sp_hyper_nonlinear_rmse[-1]
-    final_error_hyper_original = settle[-1] - sp_hyper_original_rmse[-1]
+    final_error_step = np.abs(settle[-1] - sp_step_rmse[-1])
+    final_error_hyper_nonlinear = np.abs(settle[-1] - sp_hyper_nonlinear_rmse[-1])
+    final_error_hyper_original = np.abs(settle[-1] - sp_hyper_original_rmse[-1])
 
     # (최종 계측 침하량 - 예측 침하량) 출력 (단계, 비선형 쌍곡선, 기존 쌍곡선)
     if print_values:
-        print("Error in Final Settlement (Nonlinear Hyper + Step): %0.3f" %final_error_step)
-        print("Error in Final Settlement (Nonlinear Hyperbolic): %0.3f" %final_error_hyper_nonlinear)
-        print("Error in Final Settlement (Original Hyperbolic): %0.3f" %final_error_hyper_original)
-
-
+        print("Error in Final Settlement (Nonlinear Hyper + Step): %0.3f" % final_error_step)
+        print("Error in Final Settlement (Nonlinear Hyperbolic): %0.3f" % final_error_hyper_nonlinear)
+        print("Error in Final Settlement (Original Hyperbolic): %0.3f" % final_error_hyper_original)
 
     # =====================
     # Post-Processing
@@ -351,7 +336,7 @@ def run_settle_prediction(input_file, output_dir,
 
     # 그래프 크기, 서브 그래프 개수 및 비율 설정
     fig, axes = plt.subplots(2, 1, figsize=(12, 9),
-                             gridspec_kw={'height_ratios':[1,3]})
+                             gridspec_kw={'height_ratios': [1, 3]})
 
     # 성토고 그래프 표시
     axes[0].plot(time, surcharge, color='black', label='surcharge height')
@@ -364,7 +349,8 @@ def run_settle_prediction(input_file, output_dir,
 
     # 계측 및 예측 침하량 표시
     axes[1].scatter(time[0:settle.size], -settle, s=50, facecolors='white', edgecolors='black', label='measured data')
-    axes[1].plot(time[step_start_index[0]:], -sp_step[step_start_index[0]:], linestyle='-', color='blue', label='Nonlinear + Step Loading')
+    axes[1].plot(time[step_start_index[0]:], -sp_step[step_start_index[0]:], linestyle='-', color='blue',
+                 label='Nonlinear + Step Loading')
     axes[1].plot(time_hyper, -sp_hyper_nonlinear,
                  linestyle='--', color='green', label='Nonlinear Hyperbolic')
     axes[1].plot(time_hyper, -sp_hyper_original,
@@ -478,8 +464,8 @@ def run_settle_prediction(input_file, output_dir,
     plt.title(filename + ": up to %i%% data used in the final step" % final_step_predict_percent)
 
     # 그래프 저장 (SVG 및 PNG)
-    #plt.savefig(output_dir + '/' + filename +' %i percent (SVG).svg' %final_step_predict_percent, bbox_inches='tight')
-    plt.savefig(output_dir + '/' + filename +' %i percent (PNG).png' %final_step_predict_percent, bbox_inches='tight')
+    # plt.savefig(output_dir + '/' + filename +' %i percent (SVG).svg' %final_step_predict_percent, bbox_inches='tight')
+    plt.savefig(output_dir + '/' + filename + ' %i percent (PNG).png' % final_step_predict_percent, bbox_inches='tight')
 
     # 그래프 출력
     if plot_show:
@@ -492,9 +478,16 @@ def run_settle_prediction(input_file, output_dir,
     print("Settlement prediction is done for " + filename +
           " with " + str(final_step_predict_percent) + "% data usage")
 
-    # 산정 에러값 반환
+    # 단계 성토 고려 여부 표시
+    is_multi_step = True
+    if len(step_start_index) == 1:
+        is_multi_step = False
+
+    # 반환
     return [RMSE_hyper_original, RMSE_hyper_nonlinear, RMSE_step,
             final_error_hyper_original, final_error_hyper_nonlinear,
-            final_error_step]
+            final_error_step, is_multi_step]
 
-run_settle_prediction('data/1_S-8.csv', 'output', 80, 100, False, False)
+
+#run_settle_prediction('data_1/1_SP-16.csv', 'output',
+#                      80, 100, True, True)