MATLAB Central Discussions - Join the conversation!
Main Content

Results for


Updating some of my educational Livescripts to 2024a, really love the new "define a function anywhere" feature, and have a "new" idea for improving Livescripts -- support "hidden" code blocks similar to the Jupyter Notebooks functionality.
For example, I often create "complicated" plots with a bunch of ancillary items and I don't want this code exposed to the reader by default, as it might confuse the reader. For example, consider a Livescript that might read like this:
-----
Noting the similar structure of these two mappings, let's now write a function that simply maps from some domain to some other domain using change of variable.
function x = ChangeOfVariable( x, from_domain, to_domain )
x = x - from_domain(1);
x = x * ( ( to_domain(2) - to_domain(1) ) / ( from_domain(2) - from_domain(1) ) );
x = x + to_domain(1);
end
Let's see this function in action
% HIDE CELL
clear
close all
from_domain = [-1, 1];
to_domain = [2, 7];
from_values = [-1, -0.5, 0, 0.5, 1];
to_values = ChangeOfVariable( from_values, from_domain, to_domain )
to_values = 1×5
2.0000 3.2500 4.5000 5.7500 7.0000
We can plot the values of from_values and to_values, showing how they're connected to each other:
% HIDE CELL
figure
hold on
for n = 1 : 5
plot( [from_values(n) to_values(n)], [1 0], Color="k", LineWidth=1 )
end
ax = gca;
ax.YTick = [];
ax.XLim = [ min( [from_domain, to_domain] ) - 1, max( [from_domain, to_domain] ) + 1 ];
ax.YLim = [-0.5, 1.5];
ax.XGrid = "on";
scatter( from_values, ones( 5, 1 ), Marker="s", MarkerFaceColor="flat", MarkerEdgeColor="k", SizeData=120, LineWidth=1, SeriesIndex=1 )
text( mean( from_domain ), 1.25, "$\xi$", Interpreter="latex", HorizontalAlignment="center", VerticalAlignment="middle" )
scatter( to_values, zeros( 5, 1 ), Marker="o", MarkerFaceColor="flat", MarkerEdgeColor="k", SizeData=120, LineWidth=1, SeriesIndex=2 )
text( mean( to_domain ), -0.25, "$x$", Interpreter="latex", HorizontalAlignment="center", VerticalAlignment="middle" )
scaled_arrow( ax, [mean( [from_domain(1), to_domain(1) ] ) - 1, 0.5], ( 1 - 0 ) / ( from_domain(1) - to_domain(1) ), 1 )
scaled_arrow( ax, [mean( [from_domain(end), to_domain(end)] ) + 1, 0.5], ( 1 - 0 ) / ( from_domain(end) - to_domain(end) ), -1 )
text( mean( [from_domain(1), to_domain(1) ] ) - 1.5, 0.5, "$x(\xi)$", Interpreter="latex", HorizontalAlignment="center", VerticalAlignment="middle" )
text( mean( [from_domain(end), to_domain(end)] ) + 1.5, 0.5, "$\xi(x)$", Interpreter="latex", HorizontalAlignment="center", VerticalAlignment="middle" )
-----
Where scaled_arrow is some utility function I've defined elsewhere... See how a majority of the code is simply "drivel" to create the plot, clear and close? I'd like to be able to hide those cells so that it would look more like this:
-----
Noting the similar structure of these two mappings, let's now write a function that simply maps from some domain to some other domain using change of variable.
function x = ChangeOfVariable( x, from_domain, to_domain )
x = x - from_domain(1);
x = x * ( ( to_domain(2) - to_domain(1) ) / ( from_domain(2) - from_domain(1) ) );
x = x + to_domain(1);
end
Let's see this function in action
Show code cell
from_domain = [-1, 1];
to_domain = [2, 7];
from_values = [-1, -0.5, 0, 0.5, 1];
to_values = ChangeOfVariable( from_values, from_domain, to_domain )
to_values = 1×5
2.0000 3.2500 4.5000 5.7500 7.0000
We can plot the values of from_values and to_values, showing how they're connected to each other:
Show code cell
-----
Thoughts?
I recently had issues with code folding seeming to disappear and it turns out that I had unknowingly disabled the "show code folding margin" option by accident. Despite using MATLAB for several years, I had no idea this was an option, especially since there seemed to be no references to it in the code folding part of the "Preferences" menu.
It would be great if in the future, there was a warning that told you about this when you try enable/disable folding in the Preferences.
I am using 2023b by the way.
2
17%
3
12%
4
59%
6
4%
8
5%
Other (comment below)
3%
6419 votes
In the MATLAB editor, when clicking on a variable name, all the other instances of the variable name will be highlighted.
But this does not work for structure fields, which is a pity. Such feature would be quite often useful for me.
I show an illustration below, and compare it with Visual Studio Code that does it. ;-)
I am using MATLAB R2023a, sorry if it has been added to newer versions, but I didn't see it in the release notes.
Dear members, I’m currently doing research on the subject of using Generative A.I. as a digital designer. What our research group would like to know is which ethical issues have a big impact on the decisions you guys and girls make using generative A.I.
Whether you’re using A.I. or not, we would really like to know your vision and opinion about this subject. Please empty your thoughts and oppinion into your answers, we would like to get as much information as possible.
Are you currently using A.I. when doing your job? Yes, what for. No (not yet), why not?
Using A.I., would you use real information or alter names/numbers to get an answer?
What information would or wouldn’t you use? If the client is asking/ordering you to do certain things that go against your principles, would you still do it because order is order? How far would you go?
Who is responsible for the outcome of the generated content, you or the client?
Would you still feel like a product owner if it was co-developed with A.I.?
What we are looking for is that we would like to know why people do or don’t use AI in the field of design and wich ethical considerations they make. We’re just looking for general moral line of people, for example: 70% of designers don’t feel owner of a design that is generated by AI but 95% feels owner when it is co-created.
So therefore the questions we asked, we want to know the how you feel about this.
Welcome to MATLAB Central's first Ask Me Anything (AMA) session! Over the next few weeks, I look forward to addressing any questions or curiosities you might have about MATLAB, the forum, sasquatches, or whatever's on your mind. Having volunteered as a contributor to this community before joining MathWorks, I'm excited to act as a bridge between these two worlds. Let's kick things off by sharing a little-known fact about the forum’s staff contributors!
A couple of years ago, before I joined MathWorks as a developer on the Graphics and Charting team, I often wondered who were the MathWorkers with the [staff] moniker answering questions in the Answers forum. Is their MATLAB Central activity part of their day-to-day job expectations? Do they serve specific roles on some kind of community outreach team? Is their work in the forum voluntary in the same way that non-staff contributors volunteer their time?
Now that I'm on the inside, I'd like to share a secret with my fellow MATLAB users and MATLAB Central enthusiasts: with the exception of the MathWorks Support Team, staff participation in the Answers forum is completely voluntary! The staff contributions you see in the forum arise from pure intrinsic motivation to connect with users, help people out of ruts, and spread the word about our product!
For example, Steven Lord has contributed 20-150 answers per month for 9 years! Steven is a quality engineer for core MATLAB numerical functions. Cris LaPierre develops training material and has been a faithful contributor in the forum for almost 6 years! Kojiro Saito and Akira Agata have been tackling Japanese content for more than 7 years! There are many others who have inspired me as a user, and I am honored to now call colleagues: Peter Perkins, Michio, Joss Knight, Alan Weiss, Jiro Doke, Edric Ellis, and many others who deserve appreciation.
The forum's success hinges on the invaluable contributions from the majority of non-staff volunteers, whose dedication and expertise fuel our community. But I know I wasn't alone in wondering about these staff contributors, so now you're in on the secret!
I'm curious to know what other topics you're interested in learning about. Ask me anything!
Temporary print statements are often helpful during debugging but it's easy to forget to remove the statements or sometimes you may not have writing privileges for the file. This tip uses conditional breakpoints to add print statements without ever editing the file!
What are conditional breakpoints?
Conditional breakpoints allow you to write a conditional statement that is executed when the selected line is hit and if the condition returns true, MATLAB pauses at that line. Otherwise, it continues.
The Hack: use ~fprintf() as the condition
fprintf prints information to the command window and returns the size of the message in bytes. The message size will always be greater than 0 which will always evaluate as true when converted to logical. Therefore, by negating an fprintf statement within a conditional breakpoint, the fprintf command will execute, print to the command window, and evalute as false which means the execution will continue uninterupted!
How to set a conditional break point
1. Right click the line number where you want the condition to be evaluated and select "Set Conditional Breakpoint"
2. Enter a valid MATLAB expression that returns a logical scalar value in the editor dialog.
Handy one-liners
Check if a line is reached: Don't forget the negation (~) and the line break (\n)!
~fprintf('Entered callback function\n')
Display the call stack from the break point line: one of my favorites!
~fprintf('%s\n',formattedDisplayText(struct2table(dbstack)))
Inspect variable values: For scalar values,
~fprintf('v = %.5f\n', v)
Use formattedDisplayText to convert more complex data to a string
~fprintf('%s\n', formattedDisplayText(v)).
Make sense of frequent hits: In some situations such as responses to listeners or interactive callbacks, a line can be executed 100s of times per second. Incorporate a timestamp to differentiate messages during rapid execution.
~fprintf('WindowButtonDownFcn - %s\n', datetime('now'))
Closing
This tip not only keeps your code clean but also offers a dynamic way to monitor code execution and variable states without permanent modifications. Interested in digging deeper? @Steve Eddins takes this tip to the next level with his Code Trace for MATLAB tool available on the File Exchange (read more).
Summary animation
To reproduce the events in this animation:
% buttonDownFcnDemo.m
fig = figure();
tcl = tiledlayout(4,4,'TileSpacing','compact');
for i = 1:16
ax = nexttile(tcl);
title(ax,"#"+string(i))
ax.ButtonDownFcn = @axesButtonDownFcn;
xlim(ax,[-1 1])
ylim(ax,[-1,1])
hold(ax,'on')
end
function axesButtonDownFcn(obj,event)
colors = lines(16);
plot(obj,event.IntersectionPoint(1),event.IntersectionPoint(2),...
'ko','MarkerFaceColor',colors(obj.Layout.Tile,:))
end
numel(v)
6%
length(v)
13%
width(v)
14%
nnz(v)
8%
size(v, 1)
27%
sum(v > 0)
31%
2537 votes
As far as I know, the MATLAB Community (including Matlab Central and Mathworks' official GitHub repository) has always been a vibrant and diverse professional and amateur community of MATLAB users from various fields globally. Being a part of it myself, especially in recent years, I have not only benefited continuously from the community but also tried to give back by helping other users in need.
I am a senior MATLAB user from Shenzhen, China, and I have a deep passion for MATLAB, applying it in various scenarios. Due to the less than ideal job market in my current social environment, I am hoping to find a position for remote support work within the Matlab Community. I wonder if this is realistic. For instance, Mathworks has been open-sourcing many repositories in recent years, especially in the field of deep learning with typical applications across industries. I am eager to use the latest MATLAB features to implement state-of-the-art algorithms. Additionally, I occasionally contribute through GitHub issues and pull requests.
In conclusion, I am looking forward to the opportunity to formally join the Matlab Community in a remote support role, dedicating more energy to giving back to the community and making the world a better place! (If a Mathworks employer can contact me, all the better~)
We're thrilled to unveil a new feature in the MATLAB Central community: User Following.
Our community is so lucky to have many experienced MATLAB experts who generously share their knowledge and insights across different applications, including Answers, File Exchange, Discussions, Contests, or Blogs.
With the introduction of User Following feature, you can now easily track new content across different areas and engage in discussions with people you follow. Simply click the ‘Follow’ button located on their profile page to start.
Depending on your communication setting, you will receive notifications via email and/or view updates in your ‘Followed Activity’ feeds. To tailor your feed, select the ‘People’ filter and focus on activities from those you follow.
We strongly encourage you to take advantage of the User Following feature to foster learning and collaboration within our vibrant community.
Who will be the first person you choose to follow? Share your answer in the comments section below and let's inspire each other to explore new horizons together.
How long until the 'dumbest' models are smarter than your average person? Thanks for sharing this article @Adam Danz

Hello MathWorks Community,

I am excited to announce that I am currently working on a book project centered around Matrix Algebra, specifically designed for MATLAB users. This book aims to cater to undergraduate students in engineering, where Matrix Algebra serves as a foundational element.

Matrix Algebra is not only pivotal in understanding complex engineering concepts but also in applying these principles effectively in various technological solutions. MATLAB, renowned for its powerful computational capabilities, is an excellent tool to explore and implement these concepts, making it a perfect companion for this book.

As I embark on this journey to create a resource that bridges theoretical matrix algebra with practical MATLAB applications, I am looking for one or two knowledgeable individuals who have a firm grasp of both subjects. If you have experience in teaching or applying matrix algebra in engineering contexts and are familiar with MATLAB, your contribution could be invaluable.

Collaborators will help in shaping the content to ensure it is educational, engaging, and technically robust, making complex concepts accessible and applicable for students.

If you are interested in contributing to this project or know someone who might be, please reach out to discuss how we can work together to make this book a valuable resource for engineering students.

Thank you and looking forward to your participation!

ismissing( { [ ] } )
26%
ismissing( NaN )
18%
ismissing( NaT )
11%
ismissing( missing )
21%
ismissing( categorical(missing) )
9%
ismissing( { '' } ) % 2 apostrophes
16%
896 votes
I created an ellipse visualizer in #MATLAB using App Designer! To read more about it, and how it ties to the recent total solar eclipse, check out my latest blog post:
Github Repo of the app (you can open it on MATLAB Online!):
Vonny Groose
Vonny Groose
Last activity on 8 Nov 2024 at 21:03

Mari is helping Dad work.
Today, he got dressed for work to design some new dog toy-making algorithms. #nationalpetday
Transforming my furry friend into a grayscale masterpiece with MATLAB! 🐾 #MATLABPetsDay
This is Stella while waiting to see if the code works...
Hey MATLAB Community! 🌟
As we continue to explore, learn, and innovate together, it's essential to take a moment to recognize the remarkable contributions that have sparked engaging discussions, solved perplexing problems, and shared insightful knowledge in the past two weeks. Let's dive into the highlights that have made our community even more vibrant and resourceful.

Interesting Questions

Burhan Burak brings up an intriguing issue faced when running certain code in MATLAB, seeking advice on how to refactor the code to eliminate a warning message. It's a great example of the practical challenges we often encounter
Jenni asks for guidance on improving linear models to fit data points more accurately. This question highlights the common hurdles in data analysis and model fitting, sparking a conversation on best practices and methodologies.

Popular Discussions

A thought-provoking question posed by goc3 that delves into the intricacies of MATLAB's logical operations. It's a great discussion starter that tests and expands our understanding of MATLAB's behavior.
Toshiaki Takeuchi shares an insightful visualization of the demand for MATLAB jobs across different regions, based on data from LinkedIn. This post not only provides a snapshot of the job market but also encourages members to discuss trends in MATLAB's use in the industry.

From the Blogs

Mike Croucher shares his excitement and insights on two long-awaited features finally making their way into MATLAB R2024a. His post reflects the passion and persistence of our community members in enhancing MATLAB's functionality.
In this informative post, Sivylla Paraskevopoulou offers practical tips for speeding up the training of deep learning models. It's a must-read for anyone looking to optimize their deep learning workflows.
A Heartfelt Thank You 🙏
To everyone who asked a question, started a discussion, or wrote a blog post: Thank you! Your contributions are what make our community a fountain of knowledge, inspiration, and innovation. Let's keep the momentum going and continue to support each other in our journey to explore the vast universe of MATLAB.
Happy Coding!
Note: If you haven't yet, make sure to check out these highlights and add your voice to our growing community. Your insights and experiences are what make us stronger.
Go to top of page