How do you update the legend of Figure 2 and Figure 3 instead of Figure 1?

Need:

Figure 1 should have red circle data points called "Iris Data (red)" with correct legend.

Figure 2 should have blue circle data points called "Synthetic Iris Data (blue)" with correct legend.

Figure 3 should be a combination of both graphs with correct legend.

(Iris.xls data set is attached)

Here is my erroneous code that displays the wrong legends:

%% The column vector, |species|, consists of iris flowers of three
% different species, setosa, versicolor, virginica. The double matrix
% |features| consists of four types of measurements on the flowers, the length
% and width of sepals and petals in centimeters, respectively.

features = xlsread("iris.xls", 2); %sepal length, sepal width, petal length, petal width
labels = xlsread("iris.xls", 3); % flower type classes

species = cell(size(labels));
species(labels==1)={'Setosa'};
species(labels==2)={'Versicolor'};
species(labels==3)={'Virginica'};

% The names of features are stored in an array.
feature_names = ["Sepal Length", "Sepal Width", "Petal Length", "Petal Width"];

%% Features
% Use petal length (third column in |features| ) and petal width (fourth column
% in |features| ) measurements. Save these as variables PL and PW,
% respectively.
% 4 features & 3 classes

SL = features(:,1);
SW = features(:,2);

PL = features(:,3);
PW = features(:,4);


%% Recall Original Iris Data 

% Recall Iris Setosa
Class1 = features(1:50,:);  
SLClass1 = features(1:50,1);  
SWClass1 = features(1:50,2);  
PLClass1 = features(1:50,3);  
PWClass1 = features(1:50,4);  

% Recall Iris Versicolor
Class2 = features(51:100,:);
SLClass2 = features(51:100,1);
SWClass2 = features(51:100,2);
PLClass2 = features(51:100,3);
PWClass2 = features(51:100,4);

% Recall Iris Virginica
Class3 = features(101:150,:);
SLClass3= features(101:150,1);
SWClass3= features(101:150,2);
PLClass3= features(101:150,3);
PWClass3= features(101:150,4);


%% Synthetic Data
%% Class 1 - SL vs. PW

% Plot Original Class 1 - Iris SL vs. PW.
figure
plotIris = scatter(SLClass1,PWClass1,'r')
hold on;
set(plotIris,{'DisplayName'},{'Iris Data (red)'})
legend show;
title('Original Iris Data - Class 1 - Sepal Length vs. Petal Width');
xlabel('Sepal Length');
ylabel('Petal Width');
hold off;


%% Generate Random Data

% Create Random Data with 100 Additional Observations for Each Class
% Class 1 Randoms. Generate 100 extra observations. Random Data must be 100 x 4 and normally distributed.

Class1_r = randn(100,4); 
%Class1_r = randi([0 1],100,4); % Randi did not create a truly random data.
 SLClass1_r = Class1_r(:,1);
 SWClass1_r = Class1_r(:,2);
 PLClass1_r = Class1_r(:,3);
 PWClass1_r = Class1_r(:,4);
 
a = min(SLClass1_r(:));
b = max(SLClass1_r(:));
ra = 0.0886;
rb = 0.4684;
SLClass1_Normalized = (((ra-rb) * (SLClass1_r - a)) / (b - a)) + rb;
 
a = min(PWClass1_r(:));
b = max(PWClass1_r(:));
ra = -0.9200;
rb = -0.5200;
PWClass1_Normalized = (((ra-rb) * (PWClass1_r - a)) / (b - a)) + rb;
 
 
% Plot Synthetic Data - Class 1 - SL vs. PW
figure(2)
Class1_Synthetic = scatter(SLClass1_Normalized,PWClass1_Normalized,'b')
hold on;
set(plotIris,{'DisplayName'},{'Synthetic Iris Data (blue)'})
legend show
title('Synthetic Iris Data - Class 1 - Sepal Length vs. Petal Width');
xlabel('Sepal Length');
ylabel('Petal Width');



% Plot Combined Graphs -  Class 1 - SL vs. PW
figure;
plotIris = scatter(SLClass1,PWClass1,'r')
set(plotIris,{'DisplayName'},{'Iris Data (red)'})
hold on;
legend show
title('Class 1 - SL vs. PW - Original & Synthetic Data');
xlabel('Sepal Length');
ylabel('Petal Width');
Class1_Synthetic = scatter(SLClass1_Normalized,PWClass1_Normalized,'b')
set(plotIris,{'DisplayName'},{'Synthetic Iris Data (blue)'})
legend show
hold off;

Figure1 Legend Error:

Figure2 Legend Error:

Figure3 Legend Error:

Answer 1

[Kristin Contreras]

When using the set function for Figure 2 and Figure 3,I had accidentally kept referring to the same plot name for Figure 1. Here is my corrected code.

%% The column vector, |species|, consists of iris flowers of three

% different species, setosa, versicolor, virginica. The double matrix

% |features| consists of four types of measurements on the flowers, the length

% and width of sepals and petals in centimeters, respectively.



features = xlsread("iris.xls", 2); %sepal length, sepal width, petal length, petal width

labels = xlsread("iris.xls", 3); % flower type classes



species = cell(size(labels));

species(labels==1)={'Setosa'};

species(labels==2)={'Versicolor'};

species(labels==3)={'Virginica'};



% The names of features are stored in an array.

feature_names = ["Sepal Length", "Sepal Width", "Petal Length", "Petal Width"];



%% Features

% Use petal length (third column in |features| ) and petal width (fourth column

% in |features| ) measurements. Save these as variables PL and PW,

% respectively.

% 4 features & 3 classes



SL = features(:,1);

SW = features(:,2);



PL = features(:,3);

PW = features(:,4);





%% Recall Original Iris Data 



% Recall Iris Setosa

Class1 = features(1:50,:);  

SLClass1 = features(1:50,1);  

SWClass1 = features(1:50,2);  

PLClass1 = features(1:50,3);  

PWClass1 = features(1:50,4);  



% Recall Iris Versicolor

Class2 = features(51:100,:);

SLClass2 = features(51:100,1);

SWClass2 = features(51:100,2);

PLClass2 = features(51:100,3);

PWClass2 = features(51:100,4);



% Recall Iris Virginica

Class3 = features(101:150,:);

SLClass3= features(101:150,1);

SWClass3= features(101:150,2);

PLClass3= features(101:150,3);

PWClass3= features(101:150,4);





%% Synthetic Data

%% Class 1 - SL vs. PW



% Plot Original Class 1 - Iris SL vs. PW.

figure

plotIris1 = scatter(SLClass1,PWClass1,'r')

hold on;

set(plotIris1,{'DisplayName'},{'Iris Data (red)'})

legend show;

title('Original Iris Data - Class 1 - Sepal Length vs. Petal Width');

xlabel('Sepal Length');

ylabel('Petal Width');

hold off;





%% Generate Random Data



% Create Random Data with 100 Additional Observations for Each Class

% Class 1 Randoms. Generate 100 extra observations. Random Data must be 100 x 4 and normally distributed.



Class1_r = randn(100,4); 

%Class1_r = randi([0 1],100,4); % Randi did not create a truly random data.

 SLClass1_r = Class1_r(:,1);

 SWClass1_r = Class1_r(:,2);

 PLClass1_r = Class1_r(:,3);

 PWClass1_r = Class1_r(:,4);

 

a = min(SLClass1_r(:));

b = max(SLClass1_r(:));

ra = 0.0886;

rb = 0.4684;

SLClass1_Normalized = (((ra-rb) * (SLClass1_r - a)) / (b - a)) + rb;

 

a = min(PWClass1_r(:));

b = max(PWClass1_r(:));

ra = -0.9200;

rb = -0.5200;

PWClass1_Normalized = (((ra-rb) * (PWClass1_r - a)) / (b - a)) + rb;

 

 

% Plot Synthetic Data - Class 1 - SL vs. PW

figure(2)

Class1_Syntheticslpw = scatter(SLClass1_Normalized,PWClass1_Normalized,'b')

hold on;

set(Class1_Syntheticslpw,{'DisplayName'},{'Synthetic Iris Data (blue)'})

legend show

title('Synthetic Iris Data - Class 1 - Sepal Length vs. Petal Width');

xlabel('Sepal Length');

ylabel('Petal Width');







% Plot Combined Graphs -  Class 1 - SL vs. PW

figure;

plotIris1 = scatter(SLClass1,PWClass1,'r')

set(plotIris1,{'DisplayName'},{'Iris Data (red)'})

hold on;

legend show

title('Class 1 - SL vs. PW - Original & Synthetic Data');

xlabel('Sepal Length');

ylabel('Petal Width');

Class1_Synthetic_slpw = scatter(SLClass1_Normalized,PWClass1_Normalized,'b')

set(Class1_Synthetic_slpw,{'DisplayName'},{'Synthetic Iris Data (blue)'})

legend show

hold off;

Answer 2

[Mario Chiappelli]

I have always used the legend function in the following way:

legend('nameOfGraphOne','nameOfGraphTwo',etc);

I see you are implementing a legend in the same line as the plot. I believe if you eliminate that and insert a legend using the legend function, it should work.

[Kristin Contreras]

Great tip, Mario! Thank you!

Should I use the set function in order for this to work?

Answer 3

[Mario Chiappelli]

No you don't need to. Here is some example code that I used for a plot for a class project.

figure(1);
hold on
plot(data_time,data_hour,"r-");
plot(data_time2,data_hour2,'r--');
xlabel('time(s)');
ylabel('light intensity in DANA 303');
hold off
ylim([0 10000]);
title('light intensity vs time (s) in DANA 303');
legend('Marios Sensor','Devins Sensor')