Error in color/linetype argument

Please help me, I am trying to run this code:

function perf(T,logplot)
if (nargin< 2) logplot = 0; end
colors = ['m''b''r''g''c''k''y'];
lines = ['-''-''-.''--'];
markers = ['x''x''x''x''v''^''o'];
[np,ns] = size(T);
minperf = min(T,[],2);
r = zeros(np,ns);
for p = 1: np
r(p,:) = T(p,:)/minperf(p);
end
if (logplot) r = log2(r); end
max_ratio = max(max(r));
r(find(isnan(r))) = 2*max_ratio;
r = sort(r);
clf;
for s = 1: ns
[xs,ys] = stairs(r(:,s),[1:np]/np);
option = ['-' colors(s) markers(s) ];
plot(xs,ys,option,'MarkerSize',3);
hold on;
end
axis([ 0.1 1.1*max_ratio 0 1 ]);

When I running, it appears error:

perf(T,2)

Error using plot Error in color/linetype argument.

Error in perf (line 20) plot(xs,ys,option,'MarkerSize',3);

If someone knows the answer, please help me.

[Stephen23]

@Chieu Nguyen: Your code does not make sense, for several reasons:

• [] is a concatenation operator, not a "list" operator.
• writing '' does not define a separator/delimiter, but simply is the character sequence for typing the character '.

You can easily try your code and see what it produces:

>> colors = ['m''b''r''g''c''k''y']
colors = m'b'r'g'c'k'y

In this case the square brackets are totally superfluous anyway, you would get exactly the same character vector without them:

>> colors = 'm''b''r''g''c''k''y'
colors = m'b'r'g'c'k'y

But it is unlikely that this is what you want. You need to learn about the basic data types (e.g. character array, cell array), and how to use them:

https://www.mathworks.com/help/matlab/matlab_prog/fundamental-matlab-classes.html

[Chieu Nguyen]

Thanks. I am beginner, I take this code from https://www.researchgate.net/publication/301625932/download. In my research, I need draw figures to compare something. But, I know Matlab a little bit, I mimic this code for my problem.

Answer 1

[Walter Roberson]

lines = ['-''-''-.''--'];

creates a character vector that contains -'-'-.'--

You need to learn how to use cell arrays of character vectors.

Answer 2

[Steven Lord]

Depending on the release you're using, I recommend using string arrays instead.

colors = ["m", "b", "r", "g", "c", "k", "y"];
lines = ["-", "-", "-.", "--"];
markers = ["x", "x", "x", "x", "v", "^", "o"];
plot(1:10, 1:10, colors(3)+lines(4)+markers(5))

Note that this is different from concatenating the strings together using [], which will not work.

doesNotWork = plot(1:10, 1:10, [colors(3),lines(4),markers(5)])

Using + to combine the three scalar string arrays makes a new scalar string whose contents are the contents of the shorter strings concatenated, while using [] makes a larger string array each element of which is one of the original strings. The two string arrays works and doesNotWork defined below look different and have different sizes.

works = colors(3)+lines(4)+markers(5)
doesNotWork = [colors(3), lines(4), markers(5)]
whos works doesNotWork

[Chieu Nguyen]

Thank for your answer. I am the beginner, I need to use Matlab comparing Cpu time... in my research. I just know Matlab a little bit, now I am trying to use Matlab to draw the figure. I also follow your code to get my problem ("Proximal Gradient Algorithm"). I also have a question, when I use "Accelerated proximal gradient methods", Cpu time which I obtain worse than "Proximal Gradient Algorithm". I don't know why.

[Chieu Nguyen]

Sorry, I have mistaken you are Stephen Boyd.

Answer 3

[kaushik gupta]

clc;

clf;

close all;

clear all;

x1=('Enter the sequence x:');

g=x1(end:-1:1);

i=length(x1);

j=length(g);

cov=zeros(1,i+j-1);

for k=i;

for i=j;

cov(k+i-1)=cov(k+i-1)+x1(k)*g(i);

end;

end;

m=0:length(x1)-1;

subplot(2,1,1);

stem(m,x1);

grid on;

axis([-1 length(cov) min(cov)-2 max(cov)+2]);

xlabel('...');

ylabel('original sequence');

title('Autocorrelation');

[kaushik gupta]

when i was running , it appears error:

Error using stem (line 40)

Error in color/linetype argument.

Error in mathlab9_1 (line 17)

stem(y,x1);

[Walter Roberson]

The error message shows y and x1 but your posted code shows no y. You also did not show us sample inputs.