-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathload_datasets.py
157 lines (133 loc) · 4.64 KB
/
load_datasets.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
155
156
157
import numpy as np
import pandas as pd
import random
def load_dataset(dataset_name):
path = "datasets\\"+dataset_name
if dataset_name == 'pendigits':
pendigits = pd.read_csv(path + '.csv')
x = pendigits.iloc[:, 0:-1].values
y = pendigits.iloc[:, -1].values
elif dataset_name == 'vowel':
vowel = pd.read_csv(path + '.csv')
x = vowel.iloc[:, 1:-1].values
y = vowel.iloc[:, -1].values
unique_y = np.unique(y)
numeric_y = range(0,len(unique_y))
for u_y, n_y in zip(unique_y,numeric_y):
y = np.where(y == u_y, n_y, y)
elif dataset_name == 'iris':
iris = pd.read_csv(path+'.data', header=None)
iris = iris.sample(frac=1).reset_index(drop=True)
X_iris = iris.iloc[:, 0:4].values
y_iris = iris.iloc[:, 4].values
for i in range(len(y_iris)):
if(y_iris[i] == 'Iris-setosa'):
y_iris[i] = 0
elif(y_iris[i] == 'Iris-versicolor'):
y_iris[i] = 1
else:
y_iris[i] = 2
x = X_iris
y = y_iris
elif dataset_name == 'ecoli':
ecoli = pd.read_csv(path+'.data', header=None, delim_whitespace=True)
x = ecoli.iloc[:, 1:8].values
y = ecoli.iloc[:, 8].values
for i in range(len(y)):
if y[i] == 'imL':
y[i] = 0
elif y[i] == 'imU':
y[i] = 1
elif y[i] == 'pp':
y[i] = 2
elif y[i] == 'om':
y[i] = 3
elif y[i] == 'omL':
y[i] = 4
elif y[i] == 'im':
y[i] = 5
elif y[i] == 'cp':
y[i] = 6
else: y[i] = 7
elif dataset_name == 'glass':
glass = pd.read_csv(path+'.data', header=None)
x = glass.iloc[:, 1:10].values
y = glass.iloc[:, 10].values
elif dataset_name == 'pima':
pimaIndiansDiabetes = pd.read_csv(path+'.data', header=None)
x = pimaIndiansDiabetes.iloc[:, 0:8].values
y = pimaIndiansDiabetes.iloc[:, 8].values
elif dataset_name == 'sonar':
sonarAll = pd.read_csv(path+'.data', header=None)
x = sonarAll.iloc[:, 0:60].values
y = sonarAll.iloc[:, 60].values
for i in range(len(y)):
if y[i] == 'R':
y[i] = 0
elif y[i] == 'M':
y[i] = 1
elif dataset_name == 'wine':
wine = pd.read_csv(path+'.data', header=None)
x = wine.iloc[:, 1:14].values
y = wine.iloc[:, 0].values
elif dataset_name == 'soybean': #### it has 1 more feature than the project document
soybeanSmall = pd.read_csv(path + '.data', header=None)
x = soybeanSmall.iloc[:, 0:35].values
y = soybeanSmall.iloc[:, 35].values
for i in range(len(y)):
if y[i] == 'D1':
y[i] = 0
elif y[i] == 'D2':
y[i] = 1
elif y[i] == 'D3':
y[i] = 2
elif y[i] == 'D4':
y[i] = 3
elif dataset_name == 'ionosphere': #### it has 3 data fewer than doc and 2 feature more
ionosphere = pd.read_csv(path + '.data', header=None)
x = ionosphere.iloc[:, 0:-2].values
y = ionosphere.iloc[:, -1].values
for i in range(len(y)):
if y[i] == 'g':
y[i] = 0
elif y[i] == 'b':
y[i] = 1
elif dataset_name == 'balance':
balance = pd.read_csv(path+'.data', header=None)
x = balance.iloc[:, 1:].values
y = balance.iloc[:, 0].values
for i in range(len(y)):
if y[i] == 'B':
y[i] = 0
elif y[i] == 'L':
y[i] = 1
elif y[i] == 'R':
y[i] = 2
'''
elif dataset_name == 'breast':
breast = pd.read_csv(path+'.data', header=None)
x = breast.iloc[:, 0:-1].values
y = breast.iloc[:, -1].values
# removed_index = []
n = len(x)-16
for i in range(n):
if x[i,6] == '?':
x = np.delete(x, (i), axis=0)
y = np.delete(y, (i), axis=0)
# removed_index.append(i)
# else:
# x[i,6] = float(x[i,6])
# print(removed_index)
# for i in removed_index:
# x = np.delete(x,(i),axis=0)
# y = np.delete(y,(i),axis=0)
for i in range(len(x)):
x[i,6] = float(x[i,6])
for i in range(len(y)):
if y[i] == 2:
y[i] = 0
elif y[i] == 4:
y[i] = 1
'''
k = len(set(y))
return x, y, k