Code fuer Hausarbeit a-c,e

Hausarbeit/Beispielcode von mir.md

@@ -0,0 +1,84 @@
+```
+# https://towardsdatascience.com/five-regression-python-modules-that-every-data-scientist-must-know-a4e03a886853
+#   based on: https://machinelearningmastery.com/how-to-use-correlation-to-understand-the-relationship-between-variables/
+
+# pip3 install openpyxl
+import os
+import pandas as pd  # To read data
+import math as m
+import numpy as np
+import scipy as sp 
+
+from scipy import stats
+import matplotlib.pyplot as plt  # To visualize
+
+# location will help to open files in the same directory as the py-script
+__location__ = os.path.realpath(
+    os.path.join(os.getcwd(), os.path.dirname(__file__)))
+
+df = pd.read_excel(os.path.join(__location__,'Daten_Umfrage_SPSS_20211113.xlsx'))
+
+df = df.apply(pd.to_numeric, errors='coerce')          # convert non-numeric values to NaN, e.g. Header "row 1" "CodeXYZ" -> "row 1" "NaN"
+print("Dataframe (Zeilen, Spalten, ...) inkl. NaN:", df.shape)
+print(df.head(10))
+#        Code     SE01_01  SE01_02  SE02_01  SE02_02  SE03_01 
+# 0      NaN      NaN      NaN      NaN      NaN      NaN   
+# 1      NaN      6.0      4.0      7.0      4.0      5.0   
+# ...
+#df = df.dropna()                                   # CAUTION: drops every row that even contains single NaN !
+
+print(df.tail(10))
+#          Code     SE01_01  SE01_02  SE02_01  SE02_02  SE03_01  
+# 155      NaN      4.0      4.0      3.0      5.0      1.0  
+# 156      NaN      NaN      NaN      NaN      NaN      NaN 
+# (End of File)
+
+#print(df["HO_Score_Bewerbung_Gewichtet"][105:110])
+#for col in df.columns:
+        #print(col)
+
+# Calculate Mean, gew, inv
+mwHO01_Diff = df["HO01_Diff"][1:156]            # Limit to Column and row Amount
+mwHO01_Diff = mwHO01_Diff.mean(skipna=True)     # Columns arithm. mean, skipna to ignore NaN rows
+mwHO01_Diff = round(mwHO01_Diff, 2)
+normHO01_Diff = m.sqrt((mwHO01_Diff / 6)**2)    # Norm
+invHO01_Diff = 1 - normHO01_Diff                # invert
+# usw
+print("HO01_Diff Mittelwert:", mwHO01_Diff)
+print("HO01_Diff Normiert:", normHO01_Diff)
+print("HO01_Diff Invertiert:", invHO01_Diff)
+# usw
+
+# Choose Dataframe Columns and row Amount
+dfColumnX = df["SS_Score"][1:156]
+dfColumnY = df["HO_Score_Bewerbung_Gewichtet"][1:156]
+
+# Convert Dataframe Columns to Array containing the X- and Y- Values
+arrX = np.asarray(dfColumnX)                         # convert that dataframe column to numpy array
+arrY = np.asarray(dfColumnY)                         
+
+# Prepare Plot Image
+plt.xlabel('SS_Score', color='black')
+plt.ylabel('HO_Score_Bewerbung_Gewichtet', color='black')
+plt.xlim([0,50])                                     # set x-Axis View Range,[from,to]
+plt.scatter(arrX, arrY)
+
+arrX, arrY = zip(*sorted(zip(arrX,arrY)))            # sort 2 arrays in sync 
+
+# Convert again, as sorting seemed to break the numpy array data format
+arrX = np.asarray(arrX)                              # before: "1   16.0" after: "[16. 18. 21. ...]"
+arrY = np.asarray(arrY)
+
+# Use least Square Linear Regression from SciPy Stats
+regr_results = sp.stats.linregress(arrX, arrY)   
+
+steigung = regr_results.slope
+yAchsAbschn = regr_results.intercept
+arrYpredicted = steigung * arrX + yAchsAbschn        # using y = m*x + n, calculate every single Y-Value fitting the regression Lines X-Values
+
+print("y =", steigung, "* x +", yAchsAbschn)
+
+# Plot Linear Regression Line
+plt.plot(arrX, arrYpredicted, label='Lin Regr', color='red', linestyle='solid') # https://scriptverse.academy/tutorials/python-matplotlib-plot-straight-line.html
+plt.show()
+```

Hausarbeit/UnivariateAnalyse.csv

@@ -0,0 +1,9 @@
+,battery_power,int_memory,ram
+count,200.0,200.0,200.0
+mean,1264.56,33.485,2153.125
+std,441.5502229610862,17.79559470640905,1140.4263724281582
+min,504.0,2.0,263.0
+20%,857.6,16.0,870.8000000000001
+50%,1249.5,33.0,2172.5
+80%,1721.4000000000003,51.0,3317.600000000001
+max,1999.0,64.0,3976.0

Hausarbeit/mobile_device_data.csv

@@ -0,0 +1,211 @@
+id,battery_power,bluetooth,dual_sim,4G,int_memory,ram
+1,1043,1,1,0,5,3476
+2,841,1,1,1,61,3895
+3,1807,1,0,0,27,2396
+4,1546,0,1,1,25,3893
+5,1434,0,0,1,49,1773
+6,1464,1,1,1,50,3506
+7,1718,0,0,0,47,3873
+8,833,0,1,0,62,1495
+9,1111,1,1,1,25,3485
+10,1520,0,0,0,25,651
+11,1500,0,0,0,55,3866
+12,1343,0,0,1,34,3911
+13,900,1,1,0,30,439
+14,1190,1,1,0,19,992
+15,630,0,0,1,51,2751
+16,1846,1,0,1,53,563
+17,1985,0,1,1,26,2083
+18,1042,0,0,1,48,2187
+19,1231,1,1,1,37,3902
+20,1488,0,0,0,37,2524
+21,968,0,0,1,7,1357
+22,529,0,1,0,60,3456
+23,1558,0,1,0,50,1641
+24,533,1,1,0,58,2322
+25,1037,0,1,0,5,3862
+26,1025,0,1,1,43,557
+27,1858,0,1,0,17,2427
+28,980,0,0,0,8,625
+29,644,1,1,0,15,3836
+30,1024,1,1,1,38,854
+31,1981,0,1,1,11,2383
+32,1380,0,1,1,56,926
+33,1557,1,1,0,32,3048
+34,1201,1,1,0,49,824
+35,1074,1,0,0,45,874
+36,1175,1,0,0,33,1617
+37,1280,0,1,1,8,3309
+38,1715,0,1,1,28,1786
+39,1165,0,0,0,40,3128
+40,567,0,0,0,2,2243
+41,1952,1,0,0,47,994
+42,822,1,1,0,43,2749
+43,685,1,1,0,12,793
+44,1388,1,1,1,17,324
+45,1972,0,0,0,48,366
+46,1411,1,0,1,57,1228
+47,1094,0,1,0,20,1413
+48,1653,0,1,1,13,2617
+49,916,0,1,0,33,1798
+50,1712,0,0,0,29,2693
+51,882,1,1,0,32,2437
+52,632,0,0,1,33,837
+53,1442,1,1,0,56,2815
+54,900,0,0,0,47,1440
+55,1630,1,1,0,20,3470
+56,1596,1,1,0,24,1251
+57,1272,1,0,1,54,3181
+58,1640,0,1,0,7,1361
+59,1111,0,1,0,15,1630
+60,1889,1,0,0,25,3689
+61,1907,1,1,0,22,3586
+62,529,0,1,0,56,3942
+63,578,0,0,1,38,1431
+64,1634,1,1,1,39,2167
+65,1533,1,1,1,22,1248
+66,660,0,0,1,47,1986
+67,1847,1,0,1,28,1378
+68,1206,1,1,0,10,2959
+69,549,1,0,1,53,1893
+70,1705,1,0,1,23,1676
+71,1366,0,1,0,36,1912
+72,1991,0,0,0,58,1947
+73,1102,1,0,1,40,2734
+74,1452,0,0,1,32,3617
+75,1810,1,1,0,63,3414
+76,1166,1,1,1,41,482
+77,881,1,1,0,6,2813
+78,1134,1,1,1,64,661
+79,1031,1,1,1,20,2546
+80,1376,1,1,1,34,3922
+81,1391,0,1,1,56,1199
+82,979,0,0,0,24,3779
+83,1075,1,0,0,36,3870
+84,968,0,1,1,61,858
+85,1999,0,1,1,15,3840
+86,1626,0,0,0,20,454
+87,942,0,1,0,28,3953
+88,1182,0,0,1,42,1633
+89,1982,1,1,0,48,3035
+90,1373,0,1,0,27,966
+91,1151,0,1,0,44,1761
+92,1650,1,1,0,51,1938
+93,1663,1,1,0,17,2820
+94,1965,1,0,1,3,305
+95,679,0,1,0,41,2838
+96,1465,0,0,0,3,2042
+97,1809,1,1,0,27,700
+98,757,1,1,1,26,2593
+99,1034,1,0,1,47,1835
+100,1119,1,0,0,23,3121
+101,559,1,1,1,24,2023
+102,1204,1,0,0,19,737
+103,1008,0,1,1,15,450
+104,1397,1,1,0,19,2928
+105,697,0,1,1,34,1694
+106,1939,1,0,0,58,2593
+107,1039,0,1,0,15,745
+108,1605,1,0,1,51,1310
+109,1094,0,1,0,34,2743
+110,769,1,1,1,30,3976
+111,861,1,1,0,40,3931
+112,504,0,1,0,63,3455
+113,1930,1,1,1,64,1533
+114,1795,1,1,0,52,3876
+115,1363,0,1,1,2,3239
+116,1376,1,0,1,38,3628
+117,1981,1,0,0,19,3663
+118,1901,0,0,0,62,1786
+119,1319,0,1,0,5,3502
+120,859,1,0,0,58,815
+121,1664,1,0,0,25,275
+122,955,0,1,0,62,3758
+123,517,0,1,0,38,2747
+124,1806,0,0,1,64,1809
+125,1348,0,0,1,50,2086
+126,1455,0,0,0,16,444
+127,1611,0,1,NaN,25,3467
+128,1573,1,1,1,24,2776
+129,557,0,0,1,23,3170
+130,1599,1,0,0,31,1234
+131,1051,1,1,0,16,305
+132,1857,1,1,0,14,1745
+133,1986,0,0,0,24,1707
+134,591,1,0,0,21,2220
+135,1140,0,1,1,56,3130
+136,923,1,0,1,10,1500
+137,1582,NaN,0,0,33,2145
+138,723,0,0,0,63,488
+139,1251,1,1,1,54,3863
+140,574,1,1,1,60,2913
+141,948,1,1,1,60,2094
+142,1571,0,1,0,44,3141
+143,564,1,0,0,33,2573
+144,1466,0,1,1,57, 
+145,597,1,1,0,16,3788
+146,895,1,1,1,9,3445
+147,1535,0,1,1,37,3241
+148,1832,1,0,0,44,2976
+149,1045,1,0,1,58,2241
+150,1483,1,0,1,61,3843
+151,976,0,0,1, ,3261
+152,1840,1,1,0,46,3264
+153,1807,0,1,0,8,826
+154,624,1,1,0,30,1314
+155,1963,0,0,1,38,2699
+156,1307,1,0,1,4,2565
+157,1933,1,1, ,23,3973
+158,1496,1,1,0,42,3537
+159,1532,1,0,0,7,3449
+160,1004,0,1,1,8,3895
+161,1483,1,1,1,38,2777
+162,945,0,0,0,47,2904
+163,1081,NaN,0,1,17,3774
+164,1012,0,1,1,32,3034
+165,1762,0,0,0,50,2940
+166,796,1,1,1,36, 
+167,1547,0,0,1,50,1168
+168,988,0,0,1,12,376
+169,1180,1,0,1,43,3510
+170,852,1,0,1,3,542
+171,607,1,0,1,18,550
+172,1765,0,0,0,24,791
+173,1250,0,0,0,63,1895
+174,1577,0,1,0,55,609
+175,1153,1,1,0,14,263
+176,651,1,0,0, ,2141
+177,1186,1,1,0,25,1270
+178,1429,1,0,0,33,403
+179,556,0,0,1,10,2040
+180,1735,1,0,1,21,2597
+181,1859,0,0,1,42,714
+182,915,0,0,0,10,423
+183,890,NaN,0,0,9,1238
+184,758,0,0,0,48,926
+185,541,0,0,1,28,2704
+186,586,0,1,1,33,2817
+187,762,1,1,1,39,2271
+188,683,0,1,1,9,1513
+189,1526,0,1,0,39,339
+190,1771,1,0,0,39,794
+191,639,0,0,0,28,391
+192,1783,1,0,1,23,541
+193,1933,0,0,1,55,648
+194,1384,,1,1,62,3386
+195,1770,0,0,0,26,1651
+196,1202,0,0,0,49,1177
+197,885,0,1,0,3,2887
+198,1629,1,0,1,2,2178
+199,1072,1,1,1,4,2878
+200,1863,1,1,0,64,3201
+201,1739,0,1,0,51,1490
+202,895,1,0,0,23,2724
+203,1278,1,1,0,56,3032
+204,562,1,1,1,43,3352
+205,1249,1,1,0,38,3195
+206,1811,0,0,1,25,1677
+207,560,1,1,1,12,2620
+208,1773,1,0,1,61,1061
+209,1715,1,0,0,11,1018
+210,725,0,1,1,26,1370

Hausarbeit/mobile_device_data.py

@@ -0,0 +1,107 @@
+import numpy as np  # pip3 install numpy
+import pandas as pd # pip3 install pandas
+import matplotlib.pyplot as plt
+import scipy as sp 
+from scipy import stats
+
+### a) Einlesen der Quelldaten
+# Pandas-Methode read_csv() zum Einlesen nutzen, wobei die Spalte namens "id" ausgelassen wird
+gewuenschteSpalten = ["battery_power", "bluetooth", "dual_sim", "4G", "int_memory", "ram"]
+df = pd.read_csv('mobile_device_data.csv', usecols=gewuenschteSpalten)
+
+# Print inkl. Dataframe-Methode head() aufrufen
+#   Parameter n: Ausgabe der ersten 12 Zeilen (Zeile 0 bis 11 = 12 Zeilen)
+print(df.head(n=12))
+# Ausgabe:
+#    battery_power  bluetooth  dual_sim 4G int_memory   ram
+# 0            1043        1.0         1  0          5  3476
+# ...		...			...			...	...		...		...
+# 11           1343        0.0         0  1         34  3911
+
+### b) Bereinigen
+# Dataframe enthält verschiedene Arten nicht auswertbarer Zellen:
+# 1. Zellen in denen vorher schon NaN (NotANumber) steht
+# 2. Zellen, in  denen ein leerer String steht (' ') 
+#   -> umwandeln von (' ') in "NaN", per DataFrame-Methode replace()
+df.replace(to_replace=' ', value=np.nan, inplace=True)
+
+# 1. und 2. lassen sich jetzt auf die gleiche Weise wie folgt ausgeben,
+# und es werden 4 Zeilen gefunden, in denen das Bluetooth Feld "NaN" ist,
+# und insgesamt 6 Zeilen, in denen das 4G, int_memory und ram Feld "NaN" sind:
+# print(df.isna().sum())
+# Ausgabe:
+# battery_power    0
+# bluetooth        4
+# dual_sim         0
+# 4G               2
+# int_memory       2
+# ram              2
+
+# Dataframe-Methode "dropna" aufrufen, um solche Einträge zu löschen:
+#   Parameter axis=0        : Zeile wird gelöscht
+#   Parameter how='any'     : Zum Löschen genügt eine einzelne leere Zelle
+#   Parameter inplace=True  : Die Operation wird direkt auf das DataFrame angewendet
+df.dropna(axis=0, how='any', inplace=True)
+
+# Nach dieser Änderung sind nur noch 200 Zeilen im DataFrame
+print(len(df.index))
+# Ausgabe:
+# 200
+
+### c) Univariate Analyse
+# Die Datentypen der gewünschten Merkmale werden nicht von vornherein alle als Ganzzahlen (Integer) 
+# interpretiert (vgl. print(df.info()) ), daher erst in solche umwandeln. 
+# Wenn man das nicht macht, kann .describe() nicht ordentlich mit gemischten Spaltentypen umgehen. 
+# Die Beispielausgabe wäre sonst: "mean      1264.560000        NaN   NaN"
+dfMetrischeMerkmale = df[["battery_power", "int_memory", "ram"]].astype(int)
+dfUnivariateAnalyse = dfMetrischeMerkmale[["battery_power", "int_memory", "ram"]].describe(include='all', percentiles=[0.2, 0.5, 0.8])
+dfUnivariateAnalyse.to_csv('UnivariateAnalyse.csv')
+# Ausgabe:
+#       battery_power  int_memory          ram
+# count     200.000000  200.000000   200.000000
+# mean     1264.560000   33.485000  2153.125000
+# std       441.550223   17.795595  1140.426372
+# min       504.000000    2.000000   263.000000
+# 20%       857.600000   16.000000   870.800000
+# 50%      1249.500000   33.000000  2172.500000
+# 80%      1721.400000   51.000000  3317.600000
+# max      1999.000000   64.000000  3976.000000
+
+
+### d) Balkendiagramme
+dfNominaleMerkmale = df[["bluetooth", "dual_sim", "4G"]].astype(int)
+#dfNominaleSummen = dfNominaleMerkmale["bluetooth", "dual_sim", "4G"].sum()
+#print(dfNominaleMerkmale)
+#dfNominaleMerkmale.plot(kind="hist")
+#plt.show()
+
+
+
+### e) Korrellationen nach Pearson und Lineare Regression zweier Merkmale
+print(dfMetrischeMerkmale.corr(method="pearson"))
+# Ausgabe:
+#                battery_power  int_memory       ram
+# battery_power       1.000000    0.050449 -0.069141
+# int_memory          0.050449    1.000000  0.047475
+# ram                -0.069141    0.047475  1.000000
+# -> ram und battery_power korrelieren am Stärksten, wenn auch negativ:
+
+# Per Modul SciPy Stats: Methode der kleinsten Quadrate für die Lineare Regression nutzen
+werteListeX = dfMetrischeMerkmale["ram"]
+werteListeY = dfMetrischeMerkmale["battery_power"]
+regrErgebnisse = sp.stats.linregress(werteListeX, werteListeY)   
+steigung = round(regrErgebnisse.slope, 4)
+yAchsAbschn = round(regrErgebnisse.intercept, 4)
+arrYpredicted = steigung * werteListeX + yAchsAbschn        # using y = m*x + n, calculate every single Y-Value fitting the regression Lines X-Values
+print("Regressionsgleichung:", "y =", steigung, "* x +", yAchsAbschn)
+
+# Plot Linear Regression Line
+plt.plot(werteListeX, arrYpredicted, label='Lin Regression', color='red', linestyle='solid') # https://scriptverse.academy/tutorials/python-matplotlib-plot-straight-line.html
+# Show Plot Image
+plt.xlabel('ram', color='black')
+plt.ylabel('battery_power', color='black')
+#plt.xlim([0,50])                                     # set x-Axis View Range,[from,to]
+plt.scatter(werteListeX, werteListeY)
+plt.show()
+
+

Sonstiges/TENTW/auswertung.py

@@ -91,18 +91,21 @@ df["HO_Score_erklaert"] = yAchsAbschn + df["SS_Score"] * steigung
 
 # Vorlesung 6
 df["Diff_roh_erklaert"] = df["HO_Score_Bewerbung_Roh"] - df["HO_Score_erklaert"] 
+print("Diff_roh_erklaert_Mean", df["Diff_roh_erklaert"].mean())
 df.loc[df.Diff_roh_erklaert < 0, "Diff_roh_erklaert"] = 0 # Differenz soll minimal 0 sein dürfen
 
 df["H0_Final"] = df["HO_Score_Bewerbung_Gewichtet"] + df["Diff_roh_erklaert"]
 
 df["H0_Wert_Z_Wert"] =  (df["H0_Final"] - df["H0_Final"].mean()) / df["H0_Final"].std()
+print("HO_Final mean", df["H0_Final"].mean(), "HO_Final std:", df["H0_Final"].std())
+print("SS_Mean", df["SS_Score"].mean())
 
 df["H0_Wert_7er_Skala"] =  (df["H0_Wert_Z_Wert"] * 1.5) + 4
 df["H0_Wert_7er_Skala"] =  round(df["H0_Wert_7er_Skala"], 2)
 df.loc[df.H0_Wert_7er_Skala < 1, "H0_Wert_7er_Skala"] = 1 # ausreisser festlegen auf Min 1
 df.loc[df.H0_Wert_7er_Skala > 7, "H0_Wert_7er_Skala"] = 7 # ausreisser festelegen auf Max 7
 
-#df.to_csv(__location__ + '/tmpViewFile.csv', sep=";")
+df.to_csv(__location__ + '/tmpViewFile.csv', sep=";")
 
 # Open Dataframe in Webbrowser:
 def showDf(df):