File Exchange

image thumbnail

labelpoints

version 4.1 (42 KB) by Adam Danz
Automatically label points in your plot given x and y values & an array of labels.

91 Downloads

Updated 26 Oct 2017

View License

Editor's Note: This file was selected as MATLAB Central Pick of the Week

Given vectors of x-values, y-values, and a vector of label names, lablepoints.m will automatically place all labels appropriately in the plot and output their handles. This function streamlines matlab's builtin text() function and greatly increases its flexibility.
.
xpos = [1,2,3,4]; ypos = [1,4,1,5]; labels = {'Lancaster', 'Cincinnati', 'Sofia', 'Rochester'}
h = labelpoints (xpos, ypos, labels)
.
(optional) Choose where you want your label relative to the data point using initials of compass directions (N, S, E, W, NE, NW, SE, SW, Center).
h = labelpoints (xpos, ypos, labels, 'N')
.
(optional) Add (or subtract) extra space between data points and labels by using a buffer between 0:1
h = labelpoints (xpos, ypos, labels, 'N', 0.15)
.
(optional) Decide whether or not the axis limits should automatically adjust if a label falls out of the axis limits (0/1)
h = labelpoints (xpos, ypos, labels, 'N', 0.15, 1)
.
(optional) Let labelpoints.m identify and label only the outliers in your data. Several outlier methods are available (see help labelpoints)
- Identify and label outliers that are N standard deviations from the mean
- Identify and label outliers that are greater than N times the interquartile range
- Identify and label outliers that are inside or outside of a given bound
- Identify and label outliers that are high residuals from a linear fit of the data.
.
(optional) Easily place texts in a list stacked downward, upward, to the left or right.
h = labelpoints (xpos, ypos, labels, 'N, 'stacked', 'down')
.
(optional) Rotate all labels
h = labelpoints (xpos, ypos, labels, 'rotation', 45)
.
(optional) Specify the font size and color of all labels
h = labelpoints (xpos, ypos, labels, 'FontSize', 12, 'Color', 'r')
.
The help section includes many versatile examples demonstrating labeling with numbers, strings, and symbols. This function is quite flexible as the examples show. There are only three required inputs, 3 optional inputs, and several optional parameters to increase flexibility. The outlier options may require stats toolbox.

Comments and Ratings (56)

Dan Hallatt

Adam Danz

@Eko Leksono, it works with pzmap(). Here's a demo (I'm using matlab 2018a).

H = tf([2 5 1],[1 3 5]);
pzmap(H)
grid on
hold on
plot(-1, 1, 'rx')
labelpoints(-1, 1, 'Here', 'NE', 'color', 'r')

Eko Leksono

I cannot use it in s-plane plot (pzmap) :(

Adam Danz

@ed, thanks! Subscripts and superscripts are possible with labelpoints(); just follow the example below. I've been meaning to update the code so users can specify the interpreter property but here's a way to do it in the meantime.

figure; axes
h(1) = labelpoints(.2, .2, 'H_{2}O', 'center'); %subscript
h(2) = labelpoints(.4, .4, 'H^{2}O', 'center'); %superscript
set(h, 'Interpreter', 'tex')

ed

Hi, great toolbox! For later versions it would be great to be able to use subscripts (and superscripts), for instance if points should be labeled by chemical elements.

minibela

very useful! thank you!

Brilliant, great time saver!

It would be great if you could extend the function to scatter3 plots.

Carrie Sun

Dunno why this is rated anything less than 4 or 5. This is very straightforward and intuitive. Maybe this Eva needs to practice on basic Matlab before using this. Anyway, thank you for this!

Eva Skvor

Mithun

Thank you!

Works very well. However, it would be interesting to make the extra space parameter work for logarithmic plots.

YANAN ZHU

Fabio Gomez

Adam Danz

Shane S, as for the outliers, you can get those from the boxplot output which outputs a bunch of handles used in the boxplot. Follow this example below (you may need to run it more than once in case no outliers are produced)

fakedata = normrnd(20, 50, 1, 400);
bh = boxplot(fakedata);
% find the handle that stores the outliers
hOutliers = findobj(bh,'Tag','Outliers');
% Use the handle to pull the (x,y) coordinates of the outliers
labelpoints(hOutliers.XData,hOutliers.YData, hOutliers.YData, 'E', 1,1, 'outliers_N', 10)
% give labelpoints 5 stars :D

ShaneS

How can label outliers on boxplot with this , please?

Andreas B

Thank you very much Adam, I very much appreciate it!

Adam Danz

@Andreas, it's possible to do this by making the labels have multiple lines. Here's an example:

figure; plot([1 2 3],[1 2,3], 'o')
label = sprintf('Row 1\nRow 2'); %here, the \n means new row.
labelpoints(2,2, label, 'NE', 0.1)

Try is again with
label = sprintf('Row 1\nRow 2\nRow 3');

Andreas B

This gave me almost everything I needed in one function - thank you very mucht!

Just one more question for me: is it possible to stack multiple values on multiple markers? E.g., I got a marker (x,y), on which I want to show the numerical y-value and, say, the date when it was recorded in a second line. My workaround would be to call the function twice, first with the numerical value, say, to the north and then with the date to the south.

Really great function, thanks!
Is there also a version compatible with the "App Designer"?

Does exactly what I needed. Thanks a lot!!!

Adam Danz

Thanks, Simon. Unfortunately the function only considers the x and y axes for now and plots everything on the z axis at z=0. Below is a snippet of code to show an example with a 3D plot. If you manually rotate the 3D plot you'll see that the labels are in the correct (x,y) coordinates and are at z=0. It wouldn't take a lot to add a z-axis functionality but I don't have time this week to dig into that.

figure('units','normalized','outerposition',[0 0 1 1])
[X,Y,Z] = sphere(16);
x = [0.5*X(:); 0.75*X(:); X(:)];
y = [0.5*Y(:); 0.75*Y(:); Y(:)];
z = [0.5*Z(:); 0.75*Z(:); Z(:)];
scatter3(x,y,z)

select = [200, 400, 600];
hold on;
scatter3(x(select),y(select),z(select), 'markerfacecolor', 'r')
labels = strsplit(sprintf('(%.2f, %.2f)|', reshape([x(select), y(select)]', [1,6])), '|');
labelpoints(x(select), y(select), labels(1:end-1), 'N', 'fontsize', 12)
xlabel('x axis')
ylabel('y axis')

Cool function, works fine with scatter function : thanks.
A question : How to have labels in 3D with scatter 3 ? Labels are in 2D I think.
Thanks €.

perfect works, thanks a lot

Adam Danz

Hi Brad, the function has been updated to accept datetime label positions. Let me know if there are any problems since I haven't use this func much with datetime data. However, I did test it with the following code.

t = datetime(2014,6,28) + calweeks(0:9);
y = rand(1,10);
plot(t,y);
labelpoints([datetime(2014,7,12), datetime(2014,8,24)], [0.4, 0.6], 'here->', 'W')

Hi Adam, thanks for your great work on this. I would like to use your function to annotate time series plots, where the X-values are datetime (introduced in R2014b). Your function doesn't seem to support this directly:

>> labelpoints(my_datetimes,Y_values,char(my_datetimes),'S')
Error using labelpoints (line 258)
Comparison is not defined between datetime and double arrays.

I tried a few different hacks, with no success. For example:

>> labelpoints(datenum(my_datetimes),Y_values,char(my_datetimes),'S')
Error using axis (line 80)
x-axis and y-axis limits have different data types. Use xlim and ylim functions
instead.
Error in labelpoints (line 296)
a = axis/10;% I've somewhat arbitrarily divided by 10 to make 'buffer' more
sensitive

Thanks so much!

Gizem Okyay

This is great!! Thank you!

Thank You Very much! It perfectly works for me.

Chenfei Ye

Thank you for your contribution. It works well

Adam Danz

Easy. Here's a demo:

%Fake Data
x = rand(1,10);
y = rand(1,10);
labels = rand(1,10)*10;

% add string to values
strLabels = sprintf('w=%.1f|', labels);
strLabels(end) = []; %remove final separater

%split long string into cell array
cellLabels = strsplit(strLabels, '|');

%plot and labelpoints
plot(x,y, 'o')
labelpoints(x,y,cellLabels,'NE')

Issa Asrawi

I need to add w = ... at each point in my graph. Any help please?

Hyeyun Yang

Liutong

excellent!

Craig Cole

Gordon

Works for my problem. Thank you!

Baris

A N

Very nice and helpfull tool, thx!

Very good tool. Would like to be able to select only certain strings to add as labels

Warwick

very good, easy to use.

Very helpful, thank you!

I appreciate being able to select single labels after rendering, and reposition manually. Scatter points can be very close sometimes, leading to overlapping.

Highly recommended.

found it

jvc

Fabulous! Thanks very much for this very useful tool.

Very easy to drop into my code - took me 60sec to add data labels to my figure! Thanks Adam!

cmo

Annick

Amazing!! Thanks for this.

Adam Danz

Soon I will update this file with a new version that has the option of labeling only the outliers which will be calculated within the function. Check back in a couple of weeks if interested. (3-Nov-2014)

Adam Danz

Just to follow up, the subplot issue mentioned previously has been resolved and was not a problem with labelpoint.m

This function works well with subplots, too.

Leyon

Very useful. Thanks.
Have a problem using with subplot. Even with the use of the extra option its placement of labels is crazy.
Was using it with a (4,2,s) subplot of bar graphs to plot frequencies and label with probabilities.
Any help to fix the issue would be nice

Updates

4.1

Replaced use of isdatetime() with isa() so older versions of matlab won't cause error.

4.0.0.0

fixed typos.

4.0.0.0

Move definition of 'pairIDX' within the outliers conditional.

4.0.0.0

Added several new optional parameters including interpreter, axHand, fontweight, backgroundColor. Added a new outlier method, "outlier_rand". Allow to specify label colors individually. Adapted sections of code to work with datetime axes.

3.0.0.0

Provided clearer examples and corrected a small problem when stacking text with only 1 element.

3.0.0.0

All changes listed within labelpoints.m under 'change history' starting after 11/15/14.
Two additional outlier methods were added; Added a 'stacked' option to stack text in a list. Outliers were being calculated based on medians - not we use means.

1.4.0.0

Improved the 'outliers_N' optional parameter and added the automatic removal of points that have 'inf' value.
Replaced example picture.

1.3.0.0

There was an error within the optional text rotation parameter.
curr_extent is not always a cell (ie, when only 2 label is entered).

1.2.0.0

This function has added several features onto the first version but has not affected functionality of earlier versions.

1.1.0.0

Added an additional example in the help file. No changes to code.

MATLAB Release Compatibility
Created with R2016a
Compatible with any release
Platform Compatibility
Windows macOS Linux