gpucoder - Open GPU Coder app - MATLAB (original) (raw)

Main Content

Syntax

Description

gpucoder opens the GPU Coder™ app. To create a project, in the Create GPU Coder Project dialog box, enter a name for the project and select a file location. To open an existing project, click Open and select the project file.

If you have the Embedded Coder® product, the app enables Embedded Coder features when it creates a project. To disable Embedded Coder features, in the project build settings, underAdvanced, set Use Embedded Coder features to No.

gpucoder [projectname](#d126e5440) orgpucoder -open `projectname` opens the existing project named projectname.coderprj by using the GPU Coder app.

gpucoder -new [projectname](#d126e5440) creates a GPU Coder project named projectname.coderprj and opens the GPU Coder app. If the Embedded Coder product is installed, the app enables Embedded Coder features when it creates a project.

gpucoder -ecoder false -new [projectname](#d126e5440) opens the GPU Coder app creating a project named projectname.coderprj. The app creates a project with the Embedded Coder features disabled, even if the Embedded Coder product is installed.

gpucoder -build [projectname](#d126e5440) builds the existing project named projectname.coderprj.

gpucoder -tocode [projectname](#d126e5440) -script [scriptname](#d126e5465) creates a script named scriptname.m containing the equivalent MATLAB® commands for the project settings inprojectname.coderprj.

If scriptname.m exists, gpucoder overwrites it. The script:

• Creates a configuration object named cfg that contains project build configuration.
• Defines the variable inputTypes for the function input types.
• Defines the variable globalVariables for global data initial values.
• Runs the codegen command. When you run the script, the entry-point functions that are arguments tocodegen must be on the search path.

cfg, inputTypes, andglobalVariables appear in the base workspace only after you run the script.

example

gpucoder -tocode [projectname](#d126e5440) converts the existing project named projectname.prj to the equivalent list of MATLAB commands and writes them to the Command Window.

gpucoder -toconfig [projectname](#d126e5440) exports the code configuration settings stored in a GPU Coder project file to a code configuration object. Executing this command returns a code configuration object corresponding toprojectname. For more information on which code configuration object is returned for different project file settings, see Share Build Configuration Settings.

[cfg](#mw%5F4175679e-8b66-40de-bb38-14861504680c) = gpucoder("-toconfig",[projectname](#d126e5440)) returns the code configuration settings stored in a GPU Coder project file to a code configuration object. Executing this command returns a code configuration object corresponding toprojectname. For more information on which code configuration object is returned for different project file settings, see Share Build Configuration Settings.

Examples

collapse all

Convert the GPU Coder project named myGPU_project.coderprj to the MATLAB script named myGPU_script.m.

gpucoder -tocode myGPU_project -script myGPU_script.m

Input Arguments

collapse all

Name of GPU Coder project that you want to create, open, or build. The project name must not contain spaces.

Name of script that you want to create when using the-tocode option with the -script option. The script name must not contain spaces.

Output Arguments

collapse all

Code configuration object that stores the configurations in a GPU Coder project file.

Version History

Introduced in R2017b

expand all

The gpucoder function saves projects with the.coderprj extension instead of a .prj extension. If you open a GPU Coder project created in previous releases usinggpucoder, the function saves a copy of the PRJ file as a CODERPRJ file. The GPU Coder app saves changes you make to the CODERPRJ file.

When opening a code generation project from a prior release, the app does not copy parameters specific to NVIDIA® Jetson™ or NVIDIA DRIVE® platforms from the PRJ file to the CODERPRJ file. If you open a project from a prior release that uses NVIDIA Jetson or NVIDIA DRIVE, verify the hardware settings before generating code.

See Also

Apps

• GPU Coder | GPU Environment Check

Functions

• codegen | coder.checkGpuInstall

Objects

• coder.gpuEnvConfig

Topics

• GPU Programming Paradigm
• Installing Prerequisite Products
• Setting Up the Prerequisite Products
• The GPU Environment Check and Setup App
• Generate Code by Using the GPU Coder App
• Code Generation for Deep Learning Networks by Using cuDNN
• Code Generation for Deep Learning Networks by Using TensorRT