Plotting two columns from matrix.

I am trying to plot Column 4 against Column 10 as a line plot, but only when Columns 8 and 9 are between 0 +/- 0.0005. (Columns 8, 9, 10 are x, y, and z locations).

I would have done this with two if loops which would print the row if the conditions were met. This could have then been plotted. I feel as there are over 400,000 entries this is inefficient (file attached represents a smaller quantity due to size constraints). Prealloaction is not possible as I don't know the size of the final array. What would the best way to go about this?

[Walter Roberson]

In the cases where the conditions are not met, what do you want to do? Are you wanting to do a scatter plot? Are you wanting to plot a disconnected line segment for each consecutive area where the conditions are met? Are you wanting to create a single line plot as if the entries that do not met the condition were not present at all ?

[JLV]

I do not want to use the data at all where the conditions mentioned above is not met.

A scatter plot with a line of best fit will be plotted (Column4 vs Column10).

Answer 1

[per isakson]

See Indexing with Logical Values half way down the page

A starter

%%
M = dlmread('matrix.txt');
%%
is_show = abs(M(:,8))<5e-4 & abs(M(:,9))<5e-4;
%%
figure
plot( M(is_show,10), M(is_show,4), '.' ) 

[JLV]

Very useful, reminds me logic plays an important role in coding rather than if and for loops. Thanks.

A follow up here

What would be the best way to plot Column4 if:

is_show = abs(M(:,10))<5e-4

I am trying to plot M(is_show,4) against its respective x and y coordinate locations M(is_show,8) and M(is_show,9), similar to a filled 2D contour plot. I have tried scatter too with no avail. Any suggestions on plot type will be useful.

[Walter Roberson]

x = M(is_show,8);
y = M(is_show,9);
z = M(is_show,4);
Finterp = scatteredInterpolant(x, y, z);

xmin = min(M(:,8));
xmax = max(M(:,8));
ymin = min(M(:,9));
ymax = max(M(:,9));

N = 500;
xrange = linspace(xmin, xmax, N);
yrange = linspace(ymin, ymax, N);
[XGrid, YGrid] = meshgrid(xrange, yrange);
ZGrid = Finterp(XGrid, YGrid);

contourf(XGrid, YGrid, ZGrid);
xlabel('M 8');
ylabel('M 9');
zlabel('M 4');

[JLV]

I will take a look into this and the code deeper when Im back in the office.

Thanks