%% 清空環(huán)境變量

clear

clc

%% 訓(xùn)練集/測試集產(chǎn)生

% 導(dǎo)入數(shù)據(jù)

load iris_data.mat

% 隨機(jī)產(chǎn)生訓(xùn)練集和測試集

P_train = [];

T_train = [];

P_test = [];

T_test = [];

for i = 1:3

    temp_input = features((i-1)*50+1:i*50,:);

    temp_output = classes((i-1)*50+1:i*50,:);

    n = randperm(50);

    % 訓(xùn)練集——120個樣本

    P_train = [P_train temp_input(n(1:40),:)'];

    T_train = [T_train temp_output(n(1:40),:)'];

    % 測試集——30個樣本

    P_test = [P_test temp_input(n(41:50),:)'];

    T_test = [T_test temp_output(n(41:50),:)'];

end

%% 模型建立 

result_grnn = [];

result_pnn = [];

time_grnn = [];

time_pnn = [];

for i = 1:4

    for j = i:4

        p_train = P_train(i:j,:);

        p_test = P_test(i:j,:);

       %% GRNN創(chuàng)建及仿真測試

        t = cputime;

        % 創(chuàng)建網(wǎng)絡(luò)

        net_grnn = newgrnn(p_train,T_train);

        % 仿真測試

        t_sim_grnn = sim(net_grnn,p_test);

        T_sim_grnn = round(t_sim_grnn);

        t = cputime - t;

        time_grnn = [time_grnn t];

        result_grnn = [result_grnn T_sim_grnn'];

       %% PNN創(chuàng)建及仿真測試

        t = cputime;

        Tc_train = ind2vec(T_train);

        % 創(chuàng)建網(wǎng)絡(luò)

        net_pnn = newpnn(p_train,Tc_train);

        % 仿真測試

        Tc_test = ind2vec(T_test);

        t_sim_pnn = sim(net_pnn,p_test);

        T_sim_pnn = vec2ind(t_sim_pnn);

        t = cputime - t;

        time_pnn = [time_pnn t];

        result_pnn = [result_pnn T_sim_pnn'];

    end

end

%% 性能評價

% 正確率accuracy

accuracy_grnn = [];

accuracy_pnn = [];

time = [];

for i = 1:10

    accuracy_1 = length(find(result_grnn(:,i) == T_test'))/length(T_test);

    accuracy_2 = length(find(result_pnn(:,i) == T_test'))/length(T_test);

    accuracy_grnn = [accuracy_grnn accuracy_1];

    accuracy_pnn = [accuracy_pnn accuracy_2];

end

% 結(jié)果對比

result = [T_test' result_grnn result_pnn];

accuracy = [accuracy_grnn;accuracy_pnn];

time = [time_grnn;time_pnn];

%% 繪圖

figure(1)

plot(1:30,T_test,'bo',1:30,result_grnn(:,4),'r-*',1:30,result_pnn(:,4),'k:^')

grid on

xlabel('測試集樣本編號')

ylabel('測試集樣本類別')

string = {'測試集預(yù)測結(jié)果對比(GRNN vs PNN)';['正確率:' num2str(accuracy_grnn(4)*100) '%(GRNN) vs ' num2str(accuracy_pnn(4)*100) '%(PNN)']};

title(string)

legend('真實(shí)值','GRNN預(yù)測值','PNN預(yù)測值')

figure(2)

plot(1:10,accuracy(1,:),'r-*',1:10,accuracy(2,:),'b:o')

grid on

xlabel('模型編號')

ylabel('測試集正確率')

title('10個模型的測試集正確率對比(GRNN vs PNN)')

legend('GRNN','PNN')

figure(3)

plot(1:10,time(1,:),'r-*',1:10,time(2,:),'b:o')

grid on

xlabel('模型編號')

ylabel('運(yùn)行時間(s)')

title('10個模型的運(yùn)行時間對比(GRNN vs PNN)')

legend('GRNN','PNN')