-
Notifications
You must be signed in to change notification settings - Fork 0
/
main_mineracao.py
257 lines (208 loc) · 9.3 KB
/
main_mineracao.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
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
# -*- coding: utf-8 -*-
import warnings
warnings.filterwarnings('ignore')
# http://scikit-learn.org/stable/auto_examples/classification/plot_classifier_comparison.html#example-classification-plot-classifier-comparison-py
#https://ashokharnal.wordpress.com/tag/k-nearest-neighbor-classification-example-using-python/
#http://image.slidesharecdn.com/mlevaluationgladyscastillo-110224084718-phpapp02/95/lesson-3-evaluation-and-comparison-of-supervised-learning-algorithms-67-728.jpg?cb=1393107988
from pandas.tools.plotting import parallel_coordinates
from pandas.tools.plotting import radviz
import matplotlib
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np
from sklearn.cross_validation import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.neighbors import KNeighborsClassifier
from sklearn.metrics import accuracy_score
from sklearn.metrics import confusion_matrix
from sklearn.tree import DecisionTreeClassifier
from sklearn.svm import SVC
from sklearn.naive_bayes import GaussianNB
from sklearn.ensemble import RandomForestClassifier
import datetime
import time
import glob
def radvizPlot(base, classe):
plt.figure(figsize=(10,8))
ax = radviz(base, classe)
ax.legend(loc='center left', bbox_to_anchor=(0, 1), fancybox=True, ncol=2, fontsize='x-small')
plt.ylim([-2,2])
plt.show()
def coordenatesParallels(base, classe):
plt.figure(figsize=(10,8))
ax = parallel_coordinates(base, classe)
ax.legend(loc='center left', bbox_to_anchor=(0, 1), fancybox=True, ncol=2, fontsize='x-small')
plt.show()
def accuracy(base, classe, n, classificador):
media = 0
for i in range(n):
x = base.drop([classe], axis=1)
y = base[classe]
x = StandardScaler().fit_transform(x)
X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=.3)
classificador.fit(X_train, y_train)
z = classificador.predict(X_test)
media += accuracy_score(y_test, z) * 100
return media/n
def classify(train, test, classe, n, classificador):
media = 0
for i in range(n):
trainBase = train.drop([classe], axis=1)
trainClasse = train[classe]
testBase = test.drop([classe], axis=1)
testClasse = test[classe]
trainBase = StandardScaler().fit_transform(trainBase)
testBase = StandardScaler().fit_transform(testBase)
classificador.fit(trainBase, trainClasse)
z = classificador.predict(testBase)
media += accuracy_score(testClasse, z) * 100
return media/n
def determineClassific(train, test, classe, classificador):
trainBase = train.drop([classe], axis=1)
trainClasse = train[classe]
testBase = test.drop([classe], axis=1)
testClasse = test[classe]
trainBase = StandardScaler().fit_transform(trainBase)
testBase = StandardScaler().fit_transform(testBase)
classificador.fit(trainBase, trainClasse)
return classificador.predict(testBase)
def confusionMatrix(base, classe, n, classificador):
media = 0
for i in range(n):
x = base.drop([classe], axis=1)
y = base[classe]
x = StandardScaler().fit_transform(x)
X_train, X_test, y_train, y_test = train_test_split(x, y, test_size=.3)
classificador.fit(X_train, y_train)
z = classificador.predict(X_test)
media += confusion_matrix(y_test, z)
return media/n
def categoricalConversion(data):
return pd.Categorical.from_array(data).codes
def categorical(data, atributos):
frame = pd.DataFrame()
for col in atributos:
frame[col] = pd.Categorical.from_array(data[col]).codes
return frame
def categoricalRename(data, atributos):
frame = pd.DataFrame()
l = len(atributos)
rename = np.arange(1,l)
for col, name in zip(atributos, rename):
frame[name] = pd.Categorical.from_array(data[col]).codes
return frame
def monthConversion(x):
if x == 'jan':
return 1
if x == 'feb':
return 2
if x == 'mar':
return 3
if x == 'apr':
return 4
if x == 'may':
return 5
if x == 'jun':
return 6
if x == 'jul':
return 7
if x == 'aug':
return 8
if x == 'sep':
return 9
if x == 'oct':
return 10
if x == 'nov':
return 11
if x == 'dec':
return 12
def dayConversion(x):
if x == 'sun':
return 1
if x == 'mon':
return 2
if x == 'tue':
return 3
if x == 'wed':
return 4
if x == 'thu':
return 5
if x == 'fri':
return 6
if x == 'sat':
return 7
##################### Pesquisa do Perfil Empreendedor Database #####################
#pesquisa = pd.read_csv('Pesquisa - Potencial Empreendedor dos alunos de Sistemas da Informação da UFRN-CERES.csv',sep=',')
#classe = pesquisa['Qual sua intenção ao concluir o curso?']
#pesquisa = pesquisa.drop(["Trabalha? Se sim, em que?"], axis=1)
#year = datetime.date.today().year
#pesquisa['Data de nascimento'] = pesquisa['Data de nascimento'].apply(lambda x : year - time.strptime(x, "%d/%m/%Y").tm_year)
#pesquisa = pesquisa.drop(['Qual sua intenção ao concluir o curso?'], axis=1)
#categoric = categorical(pesquisa, list(pesquisa.columns.values))
#categoric['classe'] = classe
#print (accuracy(categoric, 'classe', 1000, RandomForestClassifier()))
#print (accuracy(categoric, 'classe', 1000, SVC()))
#print (accuracy(categoric, 'classe', 1000, GaussianNB()))
#print (accuracy(categoric, 'classe', 1000, KNeighborsClassifier()))
#print (accuracy(categoric, 'classe', 1000, DecisionTreeClassifier()))
#radvizPlot(categoric, 'classe')
####################### Character Database ############################
#learn = [pd.read_table(filename, header=None, delim_whitespace=True) for filename in glob.glob("CharacterDatabase/learn/*")]
#test = [pd.read_table(filename, header=None, delim_whitespace=True) for filename in glob.glob("CharacterDatabase/test/*")]
#testUser = [pd.read_table(filename, header=None, delim_whitespace=True) for filename in glob.glob("CharacterDatabase/testUser/*")]
#dataLearn = pd.concat(learn, axis=0, ignore_index=True)
#dataTest = pd.concat(test, axis=0, ignore_index=True)
#dataTestUser = pd.concat(testUser, axis=0, ignore_index=True)
#dataLearn.columns = ['classe','line', 'str', 'x1','y1','x2','y2', 'length', 'diagonal']
#dataTest.columns = ['classe','line', 'str', 'x1','y1','x2','y2', 'length', 'diagonal']
#dataTestUser.columns = ['classe','line', 'str', 'x1','y1','x2','y2', 'length', 'diagonal']
#dataLearn = dataLearn.drop('str', axis=1)
#dataTest = dataTest.drop('str', axis=1)
#dataTestUser = dataTestUser.drop('str', axis=1)
#radvizPlot(dataLearn, 'classe')
#radvizPLot(dataTest, 'classe')
#print (accuracy(dataLearn, 'classe', 1000, RandomForestClassifier()))
#print (accuracy(dataLearn, 'classe', 1000, SVC()))
#print (accuracy(dataLearn, 'classe', 1000, GaussianNB()))
#print (accuracy(dataLearn, 'classe', 1000, KNeighborsClassifier()))
#print (accuracy(dataLearn, 'classe', 1000, DecisionTreeClassifier()))
#print (classify(dataLearn, dataTest, 'classe', 10, SVC()))
#print (classify(dataLearn, dataTest, 'classe', 10, RandomForestClassifier()))
#print (classify(dataLearn, dataTest, 'classe', 10, GaussianNB()))
#print (classify(dataLearn, dataTest, 'classe', 10, KNeighborsClassifier()))
#print (classify(dataLearn, dataTest, 'classe', 10, DecisionTreeClassifier()))
#print (determineClassific(dataLearn, dataTestUser, 'classe', RandomForestClassifier()))
########################### Forestfires #########################
#forestfires = pd.read_csv('forestfires.csv',sep=',')
#forestfires['area'] = forestfires['area'] > 0
#forestfires = forestfires.drop('X', axis=1)
#forestfires = forestfires.drop('Y', axis=1)
#forestfires['month'] = forestfires['month'].apply(lambda x: monthConversion(x))
#forestfires['day'] = forestfires['day'].apply(lambda x: dayConversion(x))
#print (accuracy(forestfires, 'area', 1000, RandomForestClassifier()))
#print (accuracy(forestfires, 'area', 1000, SVC()))
#print (accuracy(forestfires, 'area', 1000, GaussianNB()))
#print (accuracy(forestfires, 'area', 1000, KNeighborsClassifier()))
#print (accuracy(forestfires, 'area', 1000, DecisionTreeClassifier()))
#radvizPlot(forestfires, 'area')
########################### parkinsons #############################
parkinsons = pd.read_csv('parkinsons.csv',sep=',')
parkinsons['status'] = parkinsons['status'] > 0
parkinsons['patient'] = categoricalConversion(parkinsons['name'].str.rsplit('_', expand=True, n=1)[0])
parkinsons['record'] = categoricalConversion(parkinsons['name'].str.rsplit('_', expand=True, n=1)[1])
parkinsons = parkinsons.drop('name', axis=1)
exemploParkinsons = pd.read_csv('exemploParkinsons.csv',sep=',')
exemploParkinsons['status'] = exemploParkinsons['status'] > 0
exemploParkinsons['patient'] = categoricalConversion(exemploParkinsons['name'].str.rsplit('_', expand=True, n=1)[0])
exemploParkinsons['record'] = categoricalConversion(exemploParkinsons['name'].str.rsplit('_', expand=True, n=1)[1])
exemploParkinsons = exemploParkinsons.drop('name', axis=1)
#print (accuracy(parkinsons, 'status', 1000, RandomForestClassifier()))
#print (accuracy(parkinsons, 'status', 1000, SVC()))
#print (accuracy(parkinsons, 'status', 1000, GaussianNB()))
#print (accuracy(parkinsons, 'status', 1000, KNeighborsClassifier()))
#print (accuracy(parkinsons, 'status', 1000, DecisionTreeClassifier()))
#print (determineClassific(parkinsons, exemploParkinsons, 'status', RandomForestClassifier()))
print (determineClassific(parkinsons, exemploParkinsons, 'status', DecisionTreeClassifier()))
#radvizPlot(parkinsons, 'status')
#coordenatesParallels(parkinsons, 'status')
#print (parkinsons)