MaxEnt/Step4_Classification.py at main · Jiang-ge/MaxEnt · GitHub

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
from sklearn import svm
from sklearn.svm import LinearSVC
from sklearn.linear_model import SGDClassifier
from sklearn.svm import NuSVC
from sklearn import tree
from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
from sklearn import naive_bayes
from sklearn.discriminant_analysis import LinearDiscriminantAnalysis
from sklearn.discriminant_analysis import QuadraticDiscriminantAnalysis
from sklearn.ensemble import RandomForestClassifier
from sklearn.ensemble import AdaBoostClassifier
from sklearn.neural_network import MLPClassifier
import random
import math
import numpy as np
from datetime import datetime

from sklearn.tree import DecisionTreeClassifier

import MyFileOperator
def mpi_entrance():
    start_time = datetime.now()
    print(start_time)
    print('MyMaxEnt')


    def readLabel(path_label):
        readfile = open(path_label, "r")
        readAllLine = readfile.readlines()
        lineNum = len(readAllLine)
        i = 0
        returnData3 = np.empty([lineNum, 1], dtype=float)
        for readline in readAllLine:
            returnData3[i, 0] = float(readline.split(', ')[5])
            i = i + 1
        readfile.close()
        return returnData3

    def readFeatureVector_16(path_r):
        for n in range(10):
            readfile = open(path_r.format(bulk=n), "r")
            readAllLine = readfile.readlines()
            lineNum = len(readAllLine)
            i = 0
            returnData = np.empty([lineNum, 17], dtype=float)
            for readline in readAllLine:
                returnData[i, 0] = float(readline.split(', ')[0])
                returnData[i, 1] = float(readline.split(', ')[1])
                returnData[i, 2] = float(readline.split(', ')[2])
                returnData[i, 3] = float(readline.split(', ')[3])
                returnData[i, 4] = float(readline.split(', ')[4])
                returnData[i, 5] = float(readline.split(', ')[5])
                returnData[i, 6] = float(readline.split(', ')[6])
                returnData[i, 7] = float(readline.split(', ')[7])
                returnData[i, 8] = float(readline.split(', ')[8])
                returnData[i, 9] = float(readline.split(', ')[9])
                returnData[i, 10] = float(readline.split(', ')[10])
                returnData[i, 11] = float(readline.split(', ')[11])
                returnData[i, 12] = float(readline.split(', ')[12])
                returnData[i, 13] = float(readline.split(', ')[13])
                returnData[i, 14] = float(readline.split(', ')[14])
                returnData[i, 15] = float(readline.split(', ')[15])
                returnData[i, 16] = float(readline.split(', ')[16])
                i = i + 1
            readfile.close()
            if n == 0:
                featureVector = returnData
            else:
                featureVector = np.vstack((featureVector, returnData))
        return featureVector


    def Classifier(X_featureVector, Y_label):
        # clf = svm.LinearSVC(dual=False, multi_class='ovr')
        # clf = tree.DecisionTreeClassifier()
        clf = RandomForestClassifier(random_state=0)
        # clf = KNeighborsClassifier(n_neighbors=3)
        # clf = naive_bayes.GaussianNB()
        # clf = LinearDiscriminantAnalysis()
        # clf = QuadraticDiscriminantAnalysis()
        # clf = AdaBoostClassifier(DecisionTreeClassifier(max_depth=10), algorithm="SAMME", n_estimators=50, learning_rate=1)
        # clf = MLPClassifier(activation='logistic',solver='adam', max_iter=300)

        clf = clf.fit(X_featureVector, Y_label)
        print(clf)
        return clf

    sampleRatio = 0.01
    path_r_train = 'Data/cutArea/MyMaxEnt/train/{bulk}/featureVector_e213.txt'
    path_r_test = 'Data/cutArea/MyMaxEnt/test/{bulk}/featureVector_e213.txt'
    print('featureVector:featureVector_e213')
    train_featureVector_16 = readFeatureVector_16(path_r_train)
    test_featureVector_16 = readFeatureVector_16(path_r_test)
    featureVector_16 = np.vstack((train_featureVector_16, test_featureVector_16))
    path_label = 'Data/cutArea/train_test/L1C2_addC1C3_train_test.txt'
    originalLabel = readLabel(path_label)
    featureVector_16_label = np.hstack((featureVector_16, originalLabel))
    # random select 1% as training sample
    featureVector_16_label_without0 = []
    for n in range(len(featureVector_16_label)):
        if featureVector_16_label[n, 17] != 0:
            featureVector_16_label_without0.append(featureVector_16_label[n, :])
    featureVector_16_label_without0 = np.array(featureVector_16_label_without0)
    road = []
    house = []
    fence = []
    treee = []
    grass = []
    power = []
    others = []
    for j in range(len(featureVector_16_label)):
        if featureVector_16_label[j, 17] == 1:
            road.append(featureVector_16_label[j, :])
        elif featureVector_16_label[j, 17] == 2:
            house.append(featureVector_16_label[j, :])
        elif featureVector_16_label[j, 17] == 3:
            fence.append(featureVector_16_label[j, :])
        elif featureVector_16_label[j, 17] == 4:
            treee.append(featureVector_16_label[j, :])
        elif featureVector_16_label[j, 17] == 6:
            grass.append(featureVector_16_label[j, :])
        elif featureVector_16_label[j, 17] == 8:
            power.append(featureVector_16_label[j, :])
        elif featureVector_16_label[j, 17] == 0:
            others.append(featureVector_16_label[j, :])
    road = np.array(road)
    road_sample = road[np.random.choice(road.shape[0], size=int(sampleRatio*len(road)), replace=False), :]
    house = np.array(house)
    house_sample = house[np.random.choice(house.shape[0], size=int(sampleRatio*len(house)), replace=False), :]
    fence = np.array(fence)
    fence_sample = fence[np.random.choice(fence.shape[0], size=int(1*sampleRatio*len(fence)), replace=False), :]
    treee = np.array(treee)
    treee_sample = treee[np.random.choice(treee.shape[0], size=int(sampleRatio*len(treee)), replace=False), :]
    grass = np.array(grass)
    grass_sample = grass[np.random.choice(grass.shape[0], size=int(sampleRatio*len(grass)), replace=False), :]
    power = np.array(power)
    power_sample = power[np.random.choice(power.shape[0], size=int(1*sampleRatio*len(power)), replace=False), :]
    sampleFeatureVector_label = np.vstack((road_sample, house_sample))
    sampleFeatureVector_label = np.vstack((sampleFeatureVector_label, fence_sample))
    sampleFeatureVector_label = np.vstack((sampleFeatureVector_label, treee_sample))
    sampleFeatureVector_label = np.vstack((sampleFeatureVector_label, grass_sample))
    sampleFeatureVector_label = np.vstack((sampleFeatureVector_label, power_sample))


    sampleFeatureVector = sampleFeatureVector_label[:, 1:17]
    sampleLabel = sampleFeatureVector_label[:, 17]
    # 得出训练模型clf_train
    clf_train = Classifier(sampleFeatureVector, sampleLabel)
    labelPredict = clf_train.predict(featureVector_16_label_without0[:, 1:17])
    originalLabel = np.array([featureVector_16_label_without0[:, 17]]).T
    label_ori_predit = np.hstack((originalLabel, np.array([labelPredict]).T))
    label_writePath = 'Data/cutArea/MyMaxEnt/'
    # MyFileOperator.write_label(label_writePath + '1SVM_labelResult_ch2.txt', label_ori_predit)
    # MyFileOperator.write_label(label_writePath + '2DT_labelResult_ch2.txt', label_ori_predit)
    MyFileOperator.write_label(label_writePath + '3RandomForests_labelResult_e213.txt', label_ori_predit)  # 修改3
    # MyFileOperator.write_label(label_writePath + '4KNN_labelResult_ch2.txt', label_ori_predit)
    # MyFileOperator.write_label(label_writePath + '5GNB_labelResult_ch2.txt', label_ori_predit)
    # MyFileOperator.write_label(label_writePath + '6LDA_labelResult_ch2.txt', label_ori_predit)
    # MyFileOperator.write_label(label_writePath + '7QDA_labelResult_ch2.txt', label_ori_predit)
    # MyFileOperator.write_label(label_writePath + '8AB_labelResult_ch2.txt', label_ori_predit)
    # MyFileOperator.write_label(label_writePath + '9MLP_labelResult_ch2.txt', label_ori_predit)
    print('sample:' + str(sampleRatio))

    correctNum = 0
    for i in range(len(originalLabel)):
        if originalLabel[i, 0] == labelPredict[i]:
            correctNum = correctNum + 1
    accu = correctNum / len(originalLabel)
    print('3correctNum='+str(correctNum)+'; accuracy='+str(accu))


    end_time = datetime.now()
    print('All done: {}'.format(end_time - start_time))
    print(end_time)


mpi_entrance()