資源簡介
基于matlab,讀取txt文本數(shù)據(jù),對(duì)文本數(shù)據(jù)進(jìn)行處理,提取特征,然后基于BP神經(jīng)網(wǎng)絡(luò)進(jìn)行數(shù)據(jù)預(yù)測。

代碼片段和文件信息
%??將輸入的數(shù)據(jù)進(jìn)行轉(zhuǎn)化為統(tǒng)計(jì)數(shù)據(jù)
close?all;clear;
%%?加載數(shù)據(jù)?
path_train?=?‘TrainData_2015.1.1_2015.2.19.txt‘;
path_test?=?‘TestData_2015.2.20_2015.2.27.txt‘;
[traind_flavortrain_time]?=?readtxt(path_train);
[testd_flavortest_time]?=?readtxt(path_test);
%%?對(duì)數(shù)據(jù)進(jìn)行預(yù)測
%?%?當(dāng)所有時(shí)刻中虛擬機(jī)為0的規(guī)格,預(yù)測的結(jié)果是這些數(shù)據(jù)也都是為0
sum_flavor1?=?sum(train_flavor);
%?sum_flavor2?=?sum(test_flavor);
%?ind_zero?=?find(sum_flavor1==0);
%?ind_pre(1:ind_zero)?=?0;
%?對(duì)不為0的虛擬機(jī)規(guī)格進(jìn)行數(shù)量預(yù)測
data_nuum?=?length(traind_flavor(:1));
traind?=?traind_flavor(1:data_nuum?-?21:);
trainl?=?traind_flavor(22:data_nuum:);
%?data_nuum2?=?length(testd_flavor(:1));
%?testd?=?testd_flavor(1:data_nuum2?-?6:);
%?testl?=?testd_flavor(7:data_nuum2:);
imp=emd(sum_flavor1);
[Aft]?=?hhspectrum(imp);
%?求瞬時(shí)頻率
%?????infreq?=?fs?*?f;
????
????%求信號(hào)的邊界譜bjp,E:對(duì)應(yīng)的振幅值
????[Ett1]=toimage(Aftlength(t));
????enery?=?E;
????E=flipud(E);
????band?=?0;???%?對(duì)應(yīng)于邊際譜值大于1的帶寬
????num?=?0;????%?用于尋找以0.3位閾值的截止位置
????for?k=1:size(E1)
????????bjp(k)=sum(E(k:));
????????if?bjp(k)>1
????????????band?=?band+1;
????????end
????????if?bjp(k)>0.5
????????????num?=?num+1;
????????end
????????if?num?==?20
????????????stop?=?k-20;
????????end
????end
????
????area?=?polyarea(1:size(E1)bjp);???????%求頻譜面積
????
????
????
%%?測試
test_out=sim(nettraind_flavor((data_nuum?-?20):data_nuum:)‘);
test_out?=?test_out‘;
test_out(test_out<10)=0;
test_out?=?round(test_out);
%%?數(shù)據(jù)評(píng)價(jià)
N?=?15;
for?i?=1:7
????s1(i)?=?sqrt(sum((sum(test_out(i:))?-?testd_flavor(i:)).^2)/N);
????s2(i)?=?sqrt(sum(test_out(i:).^2));
????s3(i)?=?sqrt(sum(testd_flavor(i:).^2));
????score(i)?=?1-?(s1/(s2+s3));
end
%?BP的效果很差
?屬性????????????大小?????日期????時(shí)間???名稱
-----------?---------??----------?-----??----
?????文件???????8459??2018-01-10?16:42??matlab_BP\data\data_2015_1.txt
?????文件??????49174??2018-01-10?16:42??matlab_BP\data\data_2015_12.txt
?????文件??????12074??2018-02-27?20:47??matlab_BP\data\data_2015_2.txt
?????文件???????8903??2018-02-10?21:16??matlab_BP\data\data_2015_3.txt
?????文件??????13885??2018-02-10?21:16??matlab_BP\data\data_2015_4.txt
?????文件??????19945??2018-02-10?21:16??matlab_BP\data\data_2015_5.txt
?????文件??????36591??2018-01-10?16:42??matlab_BP\data\data_2016_1.txt
?????文件??????33224??2018-03-13?22:04??matlab_BP\data.emf
?????文件??????18778??2018-03-13?22:04??matlab_BP\data.fig
?????文件??????35997??2018-03-13?22:07??matlab_BP\data.png
?????文件???????1815??2018-03-13?22:12??matlab_BP\prediction.m
?????文件???????1430??2018-03-13?21:27??matlab_BP\prediction_BP.m
?????文件???????1671??2018-03-13?19:38??matlab_BP\readtxt.m
?????文件???????2643??2018-01-16?14:53??matlab_BP\TestData_2015.2.20_2015.2.27.txt
?????文件??????16125??2018-01-16?14:53??matlab_BP\TrainData_2015.1.1_2015.2.19.txt
?????目錄??????????0??2018-03-13?19:52??matlab_BP\data
?????目錄??????????0??2018-03-16?11:13??matlab_BP
-----------?---------??----------?-----??----
???????????????260714????????????????????17
- 上一篇:受控自回歸滑動(dòng)積分平均模型
- 下一篇:遞歸b樣條逼近算法確定(RBA)
評(píng)論
共有 條評(píng)論