-
Notifications
You must be signed in to change notification settings - Fork 0
/
cervical_Stacked.py
319 lines (247 loc) · 10.6 KB
/
cervical_Stacked.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
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
# -*- coding: utf-8 -*-
"""
Created on Fri Aug 30 12:26:00 2019
@author: Surbhi
"""
""" ***************************************************************************
# * File Description: *
# * The contents of this script are: *
# * 1. Importing Libraries *
# * 2. Scale Data *
# * 3. Create train and test set *
# * 4. MLP_1 *
# * 5. MLP_2 *
# * 6. MLP_3 *
# * 7. Stacking *
# * AUTHORS(S): SURBHI <sur7312@gmail.com> *
# * --------------------------------------------------------------------------*
# * ************************************************************************"""
###############################################################################
# 1. Importing Libraries #
###############################################################################
import pandas as pd
import numpy as np
import seaborn as sns
from sklearn.metrics import roc_auc_score
from sklearn.model_selection import KFold
data = pd.read_csv('cervical.csv')
#sns.heatmap(data.corr())
corr = data.corr()
ax = sns.heatmap(
corr,
vmin=-1, vmax=1, center=0,
cmap=sns.diverging_palette(20, 220, n=200),
square=True
)
ax.set_xticklabels(
ax.get_xticklabels(),
rotation=45,
horizontalalignment='right'
);
des = data.describe()
inf = data.info()
null = data.isna().sum()
null = {}
c = 0
n = 0
for i in data.columns:
for j in data[i]:
if j == '?':
c = c+1
null[i] = c
c = 0
a = {key: val for key, val in null.items() if val > 0}
for i in data.columns:
for j in data.index:
if data[i][j] == '?':
data[i][j] = np.nan
else:
pass
data.isna().sum()
data = data.drop(columns = ['STDs: Time since first diagnosis', 'STDs: Time since last diagnosis'])
data = data.apply(pd.to_numeric, errors='coerce')
d = {}
for i in data:
d[i] = data[i].nunique()
from missingpy import KNNImputer
cols = list(data)
data1 = pd.DataFrame(KNNImputer().fit_transform(data))
data1.columns = cols
data1.isna().sum()
#data2 = pd.concat([data1, y], axis = 1)
data1 = data1.dropna()
data1.isna().sum()
data1['Biopsy'].value_counts()
data1['Hinselmann'].value_counts()
data1['Citology'].value_counts()
data1['Schiller'].value_counts()
data1 = data1.drop(columns = ['Hinselmann', 'Schiller','Citology'])
X = data1.iloc[:, :-1].values
y = data1.iloc[:, -1].values
###############################################################################
# 2. Scale Data #
###############################################################################
from sklearn.preprocessing import StandardScaler
sc = StandardScaler()
X = sc.fit_transform(X)
###############################################################################
# 3. Create train and test set #
###############################################################################
from sklearn.model_selection import train_test_split
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.25, random_state = 21)
###################################################################################
trainX = X_train
trainy = y_train
testX = X_test
testy = y_test
###############################################################################
# 4. MLP_1 #
###############################################################################
from keras.models import Sequential
from keras.layers import Dense
model = Sequential()
model.add(Dense(units = 50, kernel_initializer = 'uniform',
activation = 'relu', input_dim = 30))
model.add(Dense(units = 1, kernel_initializer = 'uniform',
activation = 'sigmoid'))
model.add(Dense(units = 1, kernel_initializer = 'uniform',
activation = 'sigmoid'))
model.compile(optimizer = 'adam', loss = 'binary_crossentropy',
metrics = ['accuracy'])
history = model.fit(trainX, trainy, validation_data=(testX, testy), epochs=500, verbose=0)
_, train_acc = model.evaluate(trainX, trainy, verbose=0)
_, test_acc = model.evaluate(testX, testy, verbose=0)
print('Train: %.3f, Test: %.3f' % (train_acc, test_acc))
import os
folder = 'models'
path1 = os.getcwd()
if folder not in os.listdir(path1):
os.makedirs(folder)
filename = 'models/model_' + str(0) + '.h5'
model.save(filename)
print('>Saved %s' % filename)
###############################################################################
# 5. MLP_2 #
###############################################################################
model1 = Sequential()
model1.add(Dense(units = 150, kernel_initializer = 'uniform',
activation = 'relu', input_dim =30))
model1.add(Dense(units = 1, kernel_initializer = 'uniform',
activation = 'sigmoid'))
model1.add(Dense(units = 1, kernel_initializer = 'uniform',
activation = 'sigmoid'))
model1.add(Dense(units = 1, kernel_initializer = 'uniform',
activation = 'sigmoid'))
model1.compile(optimizer = 'adam', loss = 'binary_crossentropy',
metrics = ['accuracy'])
history = model1.fit(trainX, trainy, validation_data=(testX, testy), epochs=500, verbose=0)
_, train_acc = model1.evaluate(trainX, trainy, verbose=0)
_, test_acc = model1.evaluate(testX, testy, verbose=0)
print('Train: %.3f, Test: %.3f' % (train_acc, test_acc))
filename = 'models/model_' + str(1) + '.h5'
model1.save(filename)
print('>Saved %s' % filename)
###############################################################################
# 6. MLP_3 #
###############################################################################
model2 = Sequential()
model2.add(Dense(units = 200, kernel_initializer = 'uniform',
activation = 'relu', input_dim = 30))
model2.add(Dense(units = 1, kernel_initializer = 'uniform',
activation = 'sigmoid'))
model2.add(Dense(units = 1, kernel_initializer = 'uniform',
activation = 'sigmoid'))
model2.add(Dense(units = 1, kernel_initializer = 'uniform',
activation = 'sigmoid'))
model2.add(Dense(units = 1, kernel_initializer = 'uniform',
activation = 'sigmoid'))
model2.compile(optimizer = 'adam', loss = 'binary_crossentropy',
metrics = ['accuracy'])
model2.fit(trainX, trainy, epochs=500, verbose=0)
history = model2.fit(trainX, trainy, validation_data=(testX, testy), epochs=500, verbose=0)
_, train_acc = model2.evaluate(trainX, trainy, verbose=0)
_, test_acc = model2.evaluate(testX, testy, verbose=0)
print('Train: %.3f, Test: %.3f' % (train_acc, test_acc))
filename = 'models/model_' + str(2) + '.h5'
model2.save(filename)
print('>Saved %s' % filename)
n_model = 3
from keras.models import load_model
def load_all_models(n_models):
all_models = list()
for i in range(n_models):
filename = 'models/model_' + str(i) + '.h5'
classifier = load_model(filename)
all_models.append(classifier)
print('>loaded %s' % filename)
return all_models
n_members = 3
members = load_all_models(n_members)
print('Loaded %d models' % len(members))
for classif in members:
_, acc = classif.evaluate(testX, testy, verbose=0)
print('Model Accuracy: %.3f' % acc)
###############################################################################
# 7. Stacking #
###############################################################################
from numpy import dstack
from sklearn.linear_model import LogisticRegression
from sklearn.ensemble import GradientBoostingClassifier
from sklearn.ensemble import RandomForestClassifier
from xgboost.sklearn import XGBClassifier
from sklearn.neural_network import MLPClassifier
from sklearn.metrics import accuracy_score, confusion_matrix
from sklearn.metrics import f1_score, matthews_corrcoef, roc_auc_score
import statistics
def stacked_dataset(members, inputX):
stackX = None
for classifier in members:
yhat = classifier.predict(inputX, verbose=0)
if stackX is None:
stackX = yhat
else:
stackX = dstack((stackX, yhat))
stackX = stackX.reshape((stackX.shape[0], stackX.shape[1]*stackX.shape[2]))
return stackX
def fit_stacked_model(members, inputX, inputy):
stackedX = stacked_dataset(members, inputX)
model = GradientBoostingClassifier()
model.fit(stackedX, inputy)
return model
# model = fit_stacked_model(members, trainX, trainy)
def stacked_prediction(members, model, inputX):
stackedX = stacked_dataset(members, inputX)
yhat = model.predict(stackedX)
return yhat
# yhat = stacked_prediction(members, model, testX)
accuracy = []
f1_scores = []
mcc_score = []
auc_score = []
sensitivity_score = []
specificity_score = []
cv = KFold(n_splits = 10, random_state = 42, shuffle = True)
for train_index, test_index in cv.split(X):
X_train1, X_test1, y_train1, y_test1 = X[train_index], X[test_index], y[train_index], y[test_index]
model = fit_stacked_model(members, X_train1, y_train1)
yhat = stacked_prediction(members, model, X_test1)
acc = accuracy_score(y_test1, yhat)
print(acc)
f1 = f1_score(y_test1, yhat)
mcc = matthews_corrcoef(y_test1, yhat)
auc = roc_auc_score(y_test1, yhat)
cm = confusion_matrix(y_test1, yhat)
sensitivity = cm[0,0]/(cm[0,0]+cm[0,1])
specificity = cm[1,1]/(cm[1,0]+cm[1,1])
accuracy.append(acc)
f1_scores.append(f1)
mcc_score.append(mcc)
auc_score.append(auc)
sensitivity_score.append(sensitivity)
specificity_score.append(specificity)
print("Accuracy: " + statistics.mean(accuracy).__str__())
print("AUC: " + statistics.mean(f1_scores).__str__())
print("f1: " + statistics.mean(mcc_score).__str__())
print("MCCC: " + statistics.mean(auc_score).__str__())
print("Sensitivity: " + statistics.mean(sensitivity_score).__str__())
print("Specficity: " + statistics.mean(specificity_score).__str__())