The-Preprocessor/bivariate.py at master · krototype/The-Preprocessor · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
'''Bivariate Analysis:
Helps in bivariate analysis of data
Plotting of the data'''

import matplotlib.pyplot as plt
import matplotlib
matplotlib.style.use('ggplot')
import numpy as np
import pandas as pd
from scipy.stats import chi2_contingency
import seaborn as sns

class Bivariate:

    def __init__(self,df):
        self.df=df

    def correlation_matrix(self):
        print("Do you want correlation matrix[0/1]")
        inp=input()

        if inp=="0":
            return

        corr = self.df.corr()
        sns.heatmap(corr,
                    xticklabels=corr.columns.values,
                    yticklabels=corr.columns.values)
        plt.show()


    def continuous_continuous(self,colm1,colm2):
        self.df.plot.scatter(colm1,colm2)
        plt.xlabel(colm1)
        plt.ylabel(colm2)
        plt.show()

        print("Correlation between {} and {} is ".format(colm1,colm2))
        print(self.df[colm1].corr(self.df[colm2]))

        print('''if between -0.1 to 0.1 - WEAK RELATIONSHIP
                \nif beyond -0.5 to 0.5 - STRONG LINEAR RELATIONSHIP\n''')

        input("Press ENTER to continue")
        return


    def continuous_categorical(self,colm1,colm2):
        print("Mean when grouping according to {}".format(colm2))
        print(self.df.groupby(colm2)[colm1].mean())

        self.df.groupby(colm2)[colm1].mean().plot.bar()
        plt.ylabel(colm1)
        plt.xlabel(colm2)
        plt.show()

        var = colm2
        data = pd.concat([self.df[colm1], self.df[var]], axis=1)
        f, ax = plt.subplots(figsize=(8, 6))
        fig = sns.boxplot(x=var, y=colm1, data=data)
        plt.show()

        input("Press ENTER to continue")
        return

    def categorical_categorical(self,colm1,colm2):
        stacked_plot=pd.crosstab(self.df[colm1], self.df[colm2])
        print("Cross table of {} and {}".format(colm1,colm2))
        print(pd.crosstab(self.df[colm1],self.df[colm2]))

        print("Chi 2 test :")
        print(chi2_contingency(pd.crosstab(self.df[colm1],self.df[colm2])))


        stacked_plot.plot.bar()
        plt.ylabel(colm2)
        plt.xlabel(colm1)
        plt.show()

        input("Press ENTER to continue")
        return

    def bivariate_plot(self):
        #Landing function
        next_iter="0"

        self.correlation_matrix()

        while(next_iter!="-1"):
            print("Enter colm 1 :")
            colm1=input()
            colm1 = colm1.lower()
            type_colm1 = int(input("Is it categorical[0/1] "))

            print("Enter colm 2 :")
            colm2 = input()
            colm2 = colm2.lower()
            type_colm2 = int(input("Is it categorical[0/1] "))

            if type_colm1==1 and type_colm2==0:
                try:
                    self.continuous_categorical(colm2 , colm1)
                except RuntimeError:
                    self.continuous_categorical(colm1, colm2)

            elif type_colm1==0 and type_colm2==1:
                try:
                    self.continuous_categorical(colm1 , colm2)
                except RuntimeError:
                    self.continuous_categorical(colm2, colm1)

            elif type_colm1==0 and type_colm2==0:
                self.continuous_continuous(colm1 , colm2)

            elif type_colm1==1 and type_colm2==1:
                try:
                    self.categorical_categorical(colm1 , colm2)
                except RuntimeError:
                    self.categorical_categorical(colm2, colm1)

            next_iter=input("Press -1 to exit/ Press any other character to continue")


        return self.df