-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathMachine_Learning_eating_module_data.py
154 lines (126 loc) · 5.7 KB
/
Machine_Learning_eating_module_data.py
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
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
# -*- coding: utf-8 -*-
"""
Created on Mon Mar 16 20:10:19 2020
@author: ayanca
"""
#load library
import warnings
import numpy as np
import pandas as pd
from ML_methods_eating_module import *
from normality_test import *
from sklearn.preprocessing import LabelEncoder
from sklearn.model_selection import train_test_split
import statsmodels.api as sm
from statsmodels.formula.api import ols
from DT_hyperparameter_tune import DT_tune
################################################################################################################################################
#load data
path = r"C:\Users\ayanca\.spyder-py3\obesity_paper_1\eating-health-module-dataset\ehresp_2014.csv"
headernames = ['erbmi','eusoda', 'eusnap', 'euincome2', 'eugenhth', 'erincome', 'eudietsoda',
'euffyday', 'eufdsit', 'eufastfdfrq','ertseat', 'eudrink', 'eueat',
'euexfreq', 'euexercise', 'eufastfd', 'eumeat', 'eumilk', 'eustores',
'eustreason', 'euwic']
total_col = ['erbmi','eusoda', 'eusnap', 'eugenhth', 'euincome2', 'eudietsoda',
'euffyday', 'eufdsit', 'eufastfdfrq','ertseat', 'eudrink', 'eueat',
'euexfreq', 'euexercise', 'eufastfd', 'eumeat', 'eumilk', 'eustores',
'eustreason', 'euwic', 'body_composition']
feature_cols = ['erbmi','eusoda', 'eusnap', 'eugenhth', 'euincome2', 'eudietsoda',
'euffyday', 'eufdsit', 'eufastfdfrq','ertseat', 'eudrink', 'eueat',
'euexfreq', 'euexercise', 'eufastfd', 'eumeat', 'eumilk', 'eustores',
'eustreason', 'euwic']
data = pd.read_csv(path, na_values="?", low_memory=False)
print(data.head(), data.shape)
warnings.filterwarnings('ignore')
np.seterr(divide = 'ignore')
np.seterr(all = 'ignore')
data = remove_missing(data)
################################################################################################################################################
data = data [data>-1]
data = data[headernames]
print (data)
'''
nan_value = float("NaN")
data.replace("", nan_value, inplace=True)
data.dropna(inplace=True)
print(data.columns)
'''
data = remove_missing(data)
#data_visualize(data, headernames)
corr_sort = find_feature_reduced_matrix(data.corr())
print ("Strong dependency,", corr_sort)
to_drop = find_feature_reduced_matrix(data.corr())
print ("Remove dependency,", to_drop)
data.drop(to_drop, axis=1, inplace=True)
'''
"Underweight" = 0
"Normal Weight" = 1
"Overweight" = 2
"Obese" = 3
'''
data['body_composition'] = data.apply(calculate_body_composition, axis=1)
#data['body_composition'] = data.apply(convert_status_to_description, axis=1) #for binary classification
data = data[total_col]
print (data.head(), data.shape)
#check_feature(data)
features = data.shape[1]-1
print (data.head(10))
#remove duplicate
if data.duplicated().sum()>0:
print ("found and removed...",data.duplicated().sum())
data.drop_duplicates(inplace = True)
#############column selection based on p-value###############
selected_columns = data.values
'''
#for multiclass classification only, comment it for binary classification
import statsmodels.api as sm
def backwardElimination(x, Y, sl, columns):
numVars = len(x[0])
for i in range(0, numVars):
regressor_OLS = sm.OLS(Y, x).fit()
maxVar = max(regressor_OLS.pvalues).astype(float)
if maxVar > sl:
for j in range(0, numVars - i):
if (regressor_OLS.pvalues[j].astype(float) == maxVar):
x = np.delete(x, j, 1)
columns = np.delete(columns, j)
regressor_OLS.summary()
return x, columns
p_val = 0.05
data_modeled, selected_columns = backwardElimination(data[:, :features], data[:,features], p_val, selected_columns)
#print (data_modeled)
'''
##########################ends###############################
#shuffle data
np.random.shuffle(selected_columns)
print ("Data shape after shuffling:", selected_columns.shape)
fold = 5
X, y = selected_columns[:, :features], selected_columns[:,features]
#print (X,y)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.2, random_state=33)
#ML-compare
print ("Compare Machine Learning Algorithms Consistently....................")
print ("5 fold cross validation...")
#ensemble_all_general(X, y, fold = 5)
plot_ml_model(X, y, fold)
################################################################################################################################################
#ML-DT-gini
#print ("DT.......gini.............")
#DT_gini_accuracy(X_train, X_test, y_train, y_test, feature_cols, fold)
################################################################################################################################################
#ML-DT-entropy
#print ("DT.......entropy.............")
#DT_entropy_accuracy(X_train, X_test, y_train, y_test, feature_cols, fold)
#DT_tune(X,y,fold)
#ensemble_voting_classifier(X_train, X_test, y_train, y_test, fold)
#######################################################binary classification#########################################################
#binary classification
#ML-compare
#print ("Compare Machine Learning Algorithms Consistently....................")
#print ("5 fold cross validation...")
#ensemble_all_general(X, y, fold = 5)
#plot_ml_model(X, y, fold = 5) #SVM, DTree,RF are the best models
#calibration_model_compare (X_train, y_train, X_test, y_test)
#calibration(X_train, y_train, X_test, y_test,SVC(kernel='linear'),"SVC",2)
#from sklearn.tree import DecisionTreeClassifier
#calibration(X_train, y_train, X_test, y_test,DecisionTreeClassifier(),"DTree",2)