-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathXGB.py
109 lines (80 loc) · 3.7 KB
/
XGB.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
import statistics
from sklearn.model_selection import StratifiedKFold
from sklearn.preprocessing import StandardScaler
from xgboost import XGBClassifier
import pandas as pd
import numpy as np
from sklearn.model_selection import LeaveOneOut, GridSearchCV
from sklearn.svm import SVC
from sklearn.metrics import roc_curve, auc
from sklearn.metrics import precision_recall_fscore_support
from sklearn.metrics import f1_score
f
cv = StratifiedKFold(n_splits=10, shuffle=True, random_state=42)
data_yuhou = pd.read_excel(r"xxx.xlsx", sheet_name=0, index_col=0)#Introducing the WM disconnection feature of Whole Brain ROI
X = data_yuhou.iloc[:, :-1]
#X=X[[256,257,259,260,262,263,264,267,268,269,270,285,286,287,
# 288,289,290,291,292,293,38,294,297,42,298,44,299,300,301,
# 48,49,50,302,303,304,305,306,307,66,67,68,335,81,82,83,84,
# 85,87,88,101,102,103,104,106,110,111,112,126,128,129,131,133,
# 134,135,136,137,140,141,142,146,148,149,150,189,193,195,199,216,
# 217,218,232,233,234,235,236,237,238,241,253,254,255]]
#this is the RDN network
scaler=StandardScaler()
X=scaler.fit_transform(X)
X_S=scaler.fit_transform(X)
Y = data_yuhou.iloc[:, -1]
datalabels = np.array(Y)
X = np.array(X)
X_S=np.array(X_S)
Y = np.array(Y)
# Store weighted F1 scores for each cycle
all_weighted_f1_scores = []
for train_index, test_index in cv.split(X, Y):
X_train, X_test = X[train_index], X[test_index]
y_train, y_test = Y[train_index], Y[test_index]
XGB = XGBClassifier(objective='multi:softprob')
parameters = {'max_depth': [1,2,3], 'learning_rate': np.linspace(0.3, 1.70, 5),
'min_child_weight':[7,8,9,10,11,12,13],'n_estimators': [30,40,50,60,77,90,100]#'n_estimators': np.linspace(40, 170, 20,dtype=int)
}
clf = GridSearchCV(XGB, parameters, cv=10, n_jobs=60)
clf.fit(X_train, y_train)
best_params = clf.best_params_
best_XGB = XGBClassifier(**best_params)
print("Optimal parameter combination：", clf.best_params_)
print("Optimal model accuracy：", clf.best_score_)
best_XGB.fit(X_train, y_train)
y_pred = best_XGB.predict(X_test)
f1 = f1_score(y_test, y_pred, average=None)
weighted_f1 = f1_score(y_test, y_pred, average='weighted')
all_weighted_f1_scores.append(weighted_f1)
all_weighted_f1_scores = []
for train_index, test_index in cv.split(X, Y):
X_train, X_test = X[train_index], X[test_index]
y_train, y_test = Y[train_index], Y[test_index]
XGB = XGBClassifier(objective='multi:softprob')
parameters = {
'max_depth': [1, 2, 3],
'learning_rate': np.linspace(0.3, 1.70, 5),
'min_child_weight': [7, 8, 9, 10, 11, 12, 13],
'n_estimators': [30, 40, 50, 60, 77, 90, 100]
}
clf = GridSearchCV(XGB, parameters, cv=10, n_jobs=60)
clf.fit(X_train, y_train)
best_params = clf.best_params_
best_XGB = XGBClassifier(**best_params)
print("Optimal parameter combination:", clf.best_params_)
print("Optimal model accuracy:", clf.best_score_)
best_XGB.fit(X_train, y_train)
y_pred = best_XGB.predict(X_test)
f1 = f1_score(y_test, y_pred, average=None)
weighted_f1 = f1_score(y_test, y_pred, average='weighted')
all_weighted_f1_scores.append(weighted_f1)
confidence_interval = statistics.stdev(all_weighted_f1_scores) * 1.96
mean_performance = statistics.mean(all_weighted_f1_scores)
lower_bound = mean_performance - confidence_interval
upper_bound = mean_performance + confidence_interval
print("Confidence Interval Lower Bound:", lower_bound)
print("Confidence Interval Upper Bound:", upper_bound)
print("Mean Performance:", mean_performance)
print("Confidence Interval:", confidence_interval)