knn算法的MATLAB源碼

ID:348244 · 發表于 2018-6-9 15:57

%knn 歸一化自寫
% 把數據分為測試集和訓練集
% train=[wine(1:30,:);wine(60:89,:);wine(131:160,:)];
% test=[wine(31:59,:);wine(90:130,:);wine(161:178,:)];
% train_labels=[wine_labels(1:30,:);wine_labels(60:89,:);wine_labels(131:160,:)];
% test_labels=[wine_labels(31:59,:);wine_labels(90:130,:);wine_labels(161:178,:)];
%把標簽加到數據的最后一列
wine(:,14)=wine_labels;
%按行隨機打亂數據
wine_data =wine(randperm(size(wine,1)),:);
%取前一百條數據給訓練集
train=wine_data(1:100,1:13);
train_labels=wine_data(1:100,14);
%剩下的78條給測試集
test=wine_data(101:178,1:13);
test_labels=wine_data(101:178,14);
%數據歸一化
h=max(train,[],1);
l=min(train,[],1);
for i=1:13
train(:,i)=(train(:,i)-l(i))./(h(i)-l(i));
test(:,i)=(test(:,i)-l(i))./(h(i)-l(i));
end
%[train,strc]=mapminmax(train');
%strc歸一化返回的參數 max min
%test=mapminmax.apply(test',strc);
%計算測試集與所有訓練集間的距離
all_distance=dist(test,train');
%升序排序 B為返回的索引
[A,B]=sort(all_distance,2);
%Knn K近鄰
K=3;
all_labels=train_labels(B(:,1:K));
%統計矩陣中出現的類別
b=unique(all_labels);
%統計各個類別出現的次數
c=histc(all_labels',b);
%新標簽賦值
[m,predict_labels]=max(c);
%求準確率
accuracy=1-(length(find((predict_labels-test_labels')~=0))/length(test_labels))
%畫圖
figure;
hold on;
plot(test_labels,'o');
plot(predict_labels,'*');
xlabel('測試集數據','FontSize',12);
ylabel('類別標簽','FontSize',12);
legend('實際測試集分類','預測測試集分類');
title('測試集的實際分類和預測分類圖','FontSize',12);
grid on;
%knn 歸一化使用mapminmax函數進行的
%knn
% 把數據分為測試集和訓練集
% train=[wine(1:30,:);wine(60:89,:);wine(131:160,:)];
% test=[wine(31:59,:);wine(90:130,:);wine(161:178,:)];
% train_labels=[wine_labels(1:30,:);wine_labels(60:89,:);wine_labels(131:160,:)];
% test_labels=[wine_labels(31:59,:);wine_labels(90:130,:);wine_labels(161:178,:)];
%把標簽加到數據的最后一列
wine(:,14)=wine_labels;
%按行隨機打亂數據
wine_data =wine(randperm(size(wine,1)),:);
%取前一百條數據給訓練集
train=wine_data(1:100,1:13);
train_labels=wine_data(1:100,14);
%剩下的78條給測試集
test=wine_data(101:178,1:13);
test_labels=wine_data(101:178,14);
%數據歸一化
[train,strc]=mapminmax(train');
%strc歸一化返回的參數 max min
test=mapminmax.apply(test',strc);
%計算測試集與所有訓練集間的距離
all_distance=dist(test',train);
%升序排序 B為返回的索引
[A,B]=sort(all_distance,2);
%Knn K近鄰
K=5;
all_labels=train_labels(B(:,1:K));
%統計矩陣中出現的類別
b=unique(all_labels);
%統計各個類別出現的次數
c=histc(all_labels',b);
%新標簽賦值
[~,predict_labels]=max(c);
%求準確率
accuracy=1-(length(find((predict_labels-test_labels')~=0))/length(test_labels))
%畫圖
figure;
hold on;
plot(test_labels,'o');
plot(predict_labels,'*');
xlabel('測試集數據','FontSize',12);
ylabel('類別標簽','FontSize',12);
legend('實際測試集分類','預測測試集分類');
title('測試集的實際分類和預測分類圖','FontSize',12);
grid on;

復制代碼

帳號		自動登錄	找回密碼
密碼			立即注冊

久久久久久久999_99精品久久精品一区二区爱城_成人欧美一区二区三区在线播放_国产精品日本一区二区不卡视频_国产午夜视频_欧美精品在线观看免费

knn算法的MATLAB源碼