-
Notifications
You must be signed in to change notification settings - Fork 5
/
step3-test-ip.py
166 lines (131 loc) · 5.37 KB
/
step3-test-ip.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
import numpy as np
from sklearn import svm
import pandas as pd
from sklearn.model_selection import GridSearchCV
import joblib
from kappa import kappa
from scipy.io import loadmat
from sklearn import metrics
import time
import matplotlib.pyplot as plt
import matplotlib as mpl
import h5py
#import input_data_PU
num_labeled = 5
# 加载数据
# 独热编码转换为类别标签
#train_label=np.argmax(train_label,1)
#print(train_label)
#test_label=np.argmax(test_label,1)
#print(test_label)
#print(np.unique(test_label))
################################################### KNN训练及分类 ###################################################
OA_list = [0,0,0,0,0,0,0,0,0,0]
AA_list = [0,0,0,0,0,0,0,0,0,0]
kappa_list = [0,0,0,0,0,0,0,0,0,0]
CA_list = np.zeros((10,16))
number = 10249
seed_number = [1,2,3,4,5,6,7,8,9,10]
for INDEX in range(10):
np.random.seed(seed_number[INDEX])
f = h5py.File('data/IP_feature.h5', 'r')
data = f['data'][:]
print(data.shape)
label = f['label'][:]
print(label.shape)
f.close()
# m, n = data.max(), data.min()
# data = (data - n) / (m - n)
indices = np.arange(data.shape[0])
shuffled_indices = np.random.permutation(indices)
images = data[shuffled_indices]
labels = label[shuffled_indices]
y = labels # np.array([numpy.arange(9)[l==1][0] for l in labels])
n_classes = int(y.max() + 1)
i_labeled = []
for c in range(n_classes):
i = indices[y == c][:num_labeled] # 50
i_labeled += list(i)
l_images = images[i_labeled]
l_labels = y[i_labeled]
start = time.time()
C = np.logspace(-2, 8, 11, base=2) # 2为底数，2的-2次方到2的8次方，一共11个数
gamma = np.logspace(-2, 8, 11, base=2)
parameters = {'C': C,
'gamma': gamma}
# 问题：参数设置规律？？？
clf = GridSearchCV(svm.SVC(kernel='rbf'), parameters, cv=3, refit=True)
# Refit an estimator using the best found parameters on the whole dataset.
#print(l_images, l_labels)
clf.fit(l_images, l_labels)
print("time = ", time.time() - start)
# 输出最佳参数组合
# print('随机搜索-度量记录：',clf.cv_results_) # 包含每次训练的相关信息
print('随机搜索-最佳度量值:', clf.best_score_) # 获取最佳度量值
print('随机搜索-最佳参数：', clf.best_params_) # 获取最佳度量值时的代定参数的值。是一个字典
print('随机搜索-最佳模型：', clf.best_estimator_) # 获取最佳度量时的分类器模型
# 存储结果学习模型，方便之后的调用
# joblib.dump(clf.best_estimator_, r'.\IP\IP_BEST_MODEL_' + str(INDEX + 1) + '.m')
#
# clf = joblib.load(r".\IP\IP_BEST_MODEL_" + str(INDEX + 1) + ".m")
start = time.time()
predict_label = clf.predict(data) # (42776,)
matrix = metrics.confusion_matrix(label, predict_label)
print(matrix)
print('OA = ', np.sum(np.trace(matrix)) / float(number) * 100)
OA_list[INDEX] = np.sum(np.trace(matrix)) / float(number) * 100
kappa_temp, aa_temp, ca_temp = kappa(matrix, 16)
AA_list[INDEX] = aa_temp
CA_list[INDEX] = ca_temp
kappa_list[INDEX] = kappa_temp * 100
print("kappa = ", kappa_temp * 100)
gt = loadmat('D:\hyperspectral_data\Indian_pines_gt.mat')
gt = gt['indian_pines_gt']
# 将预测的结果匹配到图像中
new_show = np.zeros((gt.shape[0], gt.shape[1]))
k = 0
for i in range(gt.shape[0]):
for j in range(gt.shape[1]):
if gt[i][j] != 0:
new_show[i][j] = predict_label[k]
new_show[i][j] += 1
k += 1
np.save("IP/IP_" + str(np.sum(np.trace(matrix)) / float(number) * 100) + "npy", new_show)
# print new_show.shape
# 展示地物
colors = ['black', 'gray', 'lime', 'cyan', 'forestgreen', 'hotpink', 'saddlebrown',
'purple', 'red', 'yellow', 'blue', 'steelblue', 'olive', 'sandybrown', 'mediumaquamarine',
'darkorange',
'whitesmoke']
cmap = mpl.colors.ListedColormap(colors)
plt.xticks([])
plt.yticks([])
plt.imshow(new_show, cmap=cmap)
plt.savefig("IP/IP_" + str(np.sum(np.trace(matrix)) / float(number)) + "_" + str(INDEX + 1) + ".png", dpi=1000) # 保存图像
#plt.show()
print("time = ", time.time() - start)
print("\n")
print("OA_list", OA_list)
print(np.array(OA_list).mean())
print(np.array(OA_list).std())
print("AA_list", AA_list)
print(np.array(AA_list).mean())
print(np.array(AA_list).std())
print("kappa_list", kappa_list)
print(np.array(kappa_list).mean())
print(np.array(kappa_list).std())
print("CA_list", CA_list)
print(np.array(CA_list).mean(axis=0))
print(np.array(CA_list).std(axis=0))
f = open('IP/IP_results.txt', 'w')
for index in range(CA_list.mean(axis=0).shape[0]):
f.write(str(np.array(CA_list).mean(axis=0)[index]) + '\n')
f.write(str(np.array(OA_list).mean()) + '\n')
f.write(str(np.array(AA_list).mean()) + '\n')
f.write(str(np.array(kappa_list).mean()) + '\n')
f.write("\n\n\n")
for index in range(CA_list.std(axis=0).shape[0]):
f.write(str(np.array(CA_list).std(axis=0)[index]) + '\n')
f.write(str(np.array(OA_list).std()) + '\n')
f.write(str(np.array(AA_list).std()) + '\n')
f.write(str(np.array(kappa_list).std()) + '\n')