Deploy Generated MATLAB Functions from Symbolic Expressions with MATLAB Compiler

This example shows how to generate a MATLAB® function from a symbolic expression and use the function to create a standalone application with MATLAB Compiler™. The target system to run the application only requires a MATLAB Runtime installation and does not require a licensed copy of MATLAB.

This example follows the steps described in Create Standalone Application from MATLAB (MATLAB Compiler) and updates the steps to generate a MATLAB function from a symbolic expression.

Generate Deployable Function from Symbolic Expression

First, create a system of differential equations with state variables x(t) and y(t), and input parameters alpha and beta.

syms x(t) y(t) alpha beta
eqs = [x - diff(x,t) == alpha*x*y; ...
       y + diff(y,t) == beta*x*y]

eqs(t) = 
$$\left(\begin{array}{c}x\left(t\right)-\frac{\partial }{\partial t}\mathrm{ }x\left(t\right)=\alpha \hspace{0.17em}x\left(t\right)\hspace{0.17em}y\left(t\right)\\ \frac{\partial }{\partial t}\mathrm{ }y\left(t\right)+y\left(t\right)=\beta \hspace{0.17em}x\left(t\right)\hspace{0.17em}y\left(t\right)\end{array}\right)$$

To solve the system of differential equations, convert it to first-order differential equations by using the odeToVectorField function.

V = odeToVectorField(eqs);

Next, convert the symbolic expression V to a MATLAB function file by using matlabFunction. The converted function in the file myODE.m can be used without Symbolic Math Toolbox™.

matlabFunction(V,File="myODE")

ans = function_handle with value:
    @myODE

Write Function in MATLAB

Write a MATLAB function named plotODESols that solves the system of differential equations using ode45. This function accepts the input parameters alpha and beta. Because these parameters might be interpreted as character vectors in your system, convert them to the double data type. Finally, this function plots the solutions of the differential equations. Save it in the same directory as myODE.m function.

type plotODESols

function plotODESols(alpha,beta)
    if ischar(alpha)
        alpha = double(string(alpha));
    end
    if ischar(beta)
        beta = double(string(beta));
    end
    tspan = linspace(0,20,200);
    [t,y] = ode45(@(t,Y) myODE(Y,alpha,beta),tspan,[50 100]);
    plot(t,y,"o-")
    xlabel("Time")
    ylabel("Population")
end

You can use this function to create and deploy standalone application using MATLAB Compiler.

Create Standalone Application Using compiler.build.standaloneApplication

Build a standalone application using a programmatic approach. Alternatively, if you want to create a standalone application package using a graphical interface, see Create Standalone Application Using Standalone Application Compiler App (MATLAB Compiler).

Build the standalone application using the compiler.build.standaloneApplication (MATLAB Compiler) function.

buildResults = compiler.build.standaloneApplication("plotODESols.m");

The buildResults object contains information on the build type, generated files, included support packages, and build options. For details, see compiler.build.Results (MATLAB Compiler).

The function generates the following files within a folder named plotODESolsstandaloneApplication in your current working directory:

• includedSupportPackages.txt — Text file that lists all support files included in the application.

• plotODESols.exe or plotODESols — Executable file that has the .exe extension if compiled on a Windows system, or no extension if compiled on Linux or macOS systems.

• run_plotODESols.sh — Shell script file that sets the library path and executes the application. This file is only generated on Linux and macOS systems.

• mccExcludedFiles.log — Log file that contains a list of any toolbox functions that were not included in the application. For information on non-supported functions, see Limitations (MATLAB Compiler).

• readme.txt — Text file that contains information on deployment prerequisites and the list of files to package for deployment.

• requiredMCRProducts.txt — Text file that contains product IDs of products required by MATLAB Runtime to run the application.

• unresolvedSymbols.txt — Text file that contains information on unresolved symbols.

To test plotODESols from MATLAB with the input parameters alpha = 0.01 and beta = 0.02, navigate to the plotODESolsstandaloneApplication folder from within the MATLAB desktop environment and execute one of the following commands based on your operating system:

Operating System	Test in MATLAB Command Window
Windows	!plotODESols 0.01 0.02
macOS	system(['./run_plotODESols.sh ',matlabroot,' 0.01 0.02']);
Linux	!./plotODESols 0.01 0.02

Create Standalone Application Installer Using compiler.package.installer

Create an installer using the buildResults object as an input argument to the compiler.package.installer function. By default, the installer is configured to download MATLAB Runtime from the web. You can modify this and specify additional options by using name-value arguments. For details, see compiler.package.installer (MATLAB Compiler).

For example, include MATLAB Runtime in the installer. The function creates a new folder named plotODESolsinstaller that contains the standalone application installer.

compiler.package.installer(buildResults,RuntimeDelivery="installer");

To install your application using an installer created by the compiler.package.installer function, see Install Deployed Application (MATLAB Compiler). For example, in Windows® platform, you can query into the plotODESolsinstaller folder location and double-click the file MyAppInstaller.exe. Complete the installation by following the instructions in the installation wizard.

Run Standalone Application

In your system command prompt, navigate to the folder containing your standalone executable.

Run plotODESols with the input arguments 0.01 and 0.02 by using one of the following commands based on your operating system:

Operating System	Command
Windows	plotODESols 0.01 0.02
macOS	Using the shell script: ./run_plotODESols.sh <MATLAB_RUNTIME_INSTALL_DIR> 0.01 0.02 Using the executable: ./plotODESols.app/Contents/macOS/plotODESols 0.01 0.02
Linux	Using the shell script: ./run_plotODESols.sh <MATLAB_RUNTIME_INSTALL_DIR> 0.01 0.02 Using the executable: ./plotODESols 0.01 0.02

NOTE: To run the application without using the shell script on Linux and macOS, you must first add MATLAB Runtime to the library path. For details, see Set MATLAB Runtime Path for Deployment (MATLAB Compiler).

The application plots the solutions of the system of differential equations.

Deploy Standalone Application on Target Machines

Copy the plotODESolsstandaloneApplication folder to a file location on all target machines where MATLAB Runtime is installed. Run the MATLAB generated standalone application on all target machines by using the executable in the plotODESolsstandaloneApplication folder.

Alternatively, you can distribute the installer located in the plotODESolsinstaller folder. This installer will install the application along with MATLAB Runtime, which allows you to run the application outside of MATLAB.

Tips

• Instead of using the compiler.build.standaloneApplication (MATLAB Compiler) function to create a standalone application, you can also use the Standalone Application Compiler (MATLAB Compiler) app to create a standalone application.

• To produce a standalone application that does not launch a Windows command shell, use compiler.build.standaloneWindowsApplication (MATLAB Compiler).

• To specify additional compilation options, you can use the mcc (MATLAB Compiler) command to create a standalone application that does not include MATLAB Runtime or an installer.