-
Notifications
You must be signed in to change notification settings - Fork 0
/
P1_predictRegion.py
117 lines (92 loc) · 4.55 KB
/
P1_predictRegion.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
import pandas as pd
import numpy as np
from sklearn import datasets
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler
from sklearn.svm import LinearSVC
from sklearn.model_selection import train_test_split
from sklearn.svm import SVC
import matplotlib.pyplot as plt
from sklearn.neighbors import KNeighborsClassifier
import matplotlib.pyplot as plt
data_id_status = pd.read_excel('2021MCMProblemC_DataSet.xlsx')
data_jingwei=data_id_status.loc[(data_id_status['Lab Status'] == 'Positive ID')|(data_id_status['Lab Status'] == 'Negative ID'),['Latitude','Longitude','Lab Status']]
data_jingwei.loc[data_jingwei['Lab Status'] == 'Positive ID','Lab Status']=1
data_jingwei.loc[data_jingwei['Lab Status'] == 'Negative ID','Lab Status']=0
data_jingwei_negativesample=data_jingwei[data_jingwei['Lab Status']==0].sample(n=14)
data_jingwei=data_jingwei[data_jingwei['Lab Status']==1].append(data_jingwei_negativesample)
data_jingwei=data_jingwei.sample(frac=1)
fig1=plt.figure()
plt.scatter(data_jingwei[data_jingwei['Lab Status']==0]['Longitude'],data_jingwei[data_jingwei['Lab Status']==0]['Latitude'],s=1,color='b',marker='^')
plt.scatter(data_jingwei[data_jingwei['Lab Status']==1]['Longitude'],data_jingwei[data_jingwei['Lab Status']==1]['Latitude'],s=20,color='r')
plt.show()
x_train=data_jingwei.loc[:,['Latitude','Longitude']].values
y_train=data_jingwei['Lab Status'].values
y_train=y_train.astype('int')
# x_train, x_test, y_train, y_test = train_test_split(x,y, random_state=1, train_size=0.6)
def plot_dataset(X,y, axes):
plt.plot( X[:,1][y==0], X[:,0][y==0], "bs" )
plt.plot( X[:,1][y==1], X[:,0][y==1], "r^" )
plt.legend(['Negative','Positive'])
# plt.axis( axes )
plt.grid( True, which="both" )
# contour函数是画出轮廓，需要给出X和Y的网格，以及对应的Z，它会画出Z的边界（相当于边缘检测及可视化）
def plot_predict(clf, axes):
x0s = np.linspace(axes[0], axes[1], 100)
x1s = np.linspace(axes[2], axes[3], 100)
x0, x1 = np.meshgrid( x0s, x1s )
X = np.c_[x0.ravel(), x1.ravel()]
y_pred = clf.predict( X ).reshape( x0.shape )
y_decision = clf.decision_function( X ).reshape( x0.shape )
plt.contour( x1, x0, y_pred, cmap=plt.cm.winter, alpha=0.5 )
C=plt.contour( x1, x0, y_decision, cmap=plt.cm.spring, alpha=0.5 )
plt.clabel(C, inline=True, fontsize=10,colors='black')
svm = Pipeline([
("scaler", StandardScaler()),
("svm_clf", SVC(kernel="rbf", gamma=1, C=1))
])
# svm=SVC(kernel="linear")
fig2=plt.figure(figsize=(6,3))
svm.fit( x_train,y_train )
plot_dataset( x_train,y_train, [45,50,-125,-115] )
plot_predict( svm, [45,50,-125,-115] )
plt.xlabel('Longitude')
plt.ylabel('Latitude')
plt.show()
# use knn to train
# 模型训练
k = 5
clf = KNeighborsClassifier(n_neighbors=7)
clf.fit(x_train, y_train)
# 进行预测
X_sample = np.array([[48.75,-122.7]])
y_sample = clf.predict(X_sample)
neighbors=clf.kneighbors(X_sample, return_distance=False)
#取出来的点是训练样本X里的索引
# 画出示意图
plt.figure(figsize=(8,5))
# plt.xlim((-116,-125))
# plt.ylim(((45,50)))
plt.scatter(data_jingwei[data_jingwei['Lab Status']==0]['Longitude'],data_jingwei[data_jingwei['Lab Status']==0]['Latitude'],s=20,color='b',marker='^')
plt.scatter(data_jingwei[data_jingwei['Lab Status']==1]['Longitude'],data_jingwei[data_jingwei['Lab Status']==1]['Latitude'],s=20,color='r')
plt.scatter(X_sample[0][1],X_sample[0][0],s=50,color='y',marker='*')
for i in neighbors[0]:
plt.plot([x_train[i][1], X_sample[0][1]], [x_train[i][0], X_sample[0][0]],
'-.', linewidth=0.6); # 预测点与距离最近的 5 个样本的连线
plt.show()
# stacking&blending
data_id_status = pd.read_excel('2021MCMProblemC_DataSet.xlsx')
data_jingwei=data_id_status.loc[(data_id_status['Lab Status'] == 'Positive ID')|(data_id_status['Lab Status'] == 'Negative ID'),['Latitude','Longitude','Lab Status']]
data_jingwei.loc[data_jingwei['Lab Status'] == 'Positive ID','Lab Status']=1
data_jingwei.loc[data_jingwei['Lab Status'] == 'Negative ID','Lab Status']=0
x_train=data_jingwei.loc[:,['Latitude','Longitude']].values
y_train=data_jingwei['Lab Status'].values
y_train=y_train.astype('int')
y_tp=svm.predict(x_train)
acc_svm=np.sum((y_tp==y_train).astype(float))/x_train.shape[0]
print(acc_svm)
y_tp=clf.predict(x_train)
acc_knn=np.sum((y_tp==y_train).astype(float))/x_train.shape[0]
print(acc_knn)
print(acc_svm/(acc_svm+acc_knn),acc_knn/(acc_knn+acc_svm))
# final goal = 0.5083783783783784 0.49162162162162165