Quickstart

If you already have the generated model code, skip to step 2. If you get the joy of generating the model source, well start from the top.

Step 1: Generate the Model’s Code.

This is likely the hardest part of the whole process. But we’ve tried to make it simple for you. You just need to check a few settings and build the model. If you’re feeling bold, you can copy-paste a script to do it for you!

Here’s said script:

model = 'MyAwesomeModel';        % <-- The name of your model...change this!
configs = getActiveConfigSet(model);
mustProp=false;
if isa(configs, 'Simulink.ConfigSetRef')
    configs = eval(configs.SourceName);
    mustProp=true;
end

set_param(configs, 'ConcurrentTasks', 'off');
set_param(configs, 'EnableMultiTasking', 'off');
set_param(configs, 'SystemTargetFile', 'grt.tlc');
set_param(configs, 'GenerateMakefile', 'off');
set_param(configs, 'GenCodeOnly', 'on');
set_param(configs, 'PackageGeneratedCodeAndArtifacts', 'on');
set_param(configs, 'RTWCAPIParams', 'on');
set_param(configs, 'RTWCAPISignals', 'on');
set_param(configs, 'RTWCAPIRootIO', 'on');
set_param(configs, 'RTWCAPIStates', 'on');
set_param(configs, 'GRTInterface', 'off');
set_param(configs, 'CombineOutputUpdateFcns', 'on');

if mustProp
    Simulink.BlockDiagram.propagateConfigSet(model);
end
slbuild(model);

What’s going on here?

It’s just changing a few settings in your model’s config. These are the minimum settings required for PySimlink, and should not affect your model’s ability to build (knock on wood)…

Want to do this manually? Check out the How-To guide for details.

Step 2: Let PySimlink Handle The Rest

At this point, you should have a copy of the model in the form of generated code. Let’s say it’s called my_awesome_model.zip and the name of your root model is “my_awesome_model”.

Tip

The zip file does not needed to be named after the model. To slightly improve build time, you can also extract the zip file and provide the path to the directory that contains the zip folder’s contents instead.

First, install PySimlink

pip install pysimlink

Now, you can import and inspect the model.

from pysimlink import Model, print_all_params

my_awesome_model = Model("my_awesome_model", "./my_awesome_model.zip")
my_awesome_mode.reset()
print_all_params(my_awesome_model)

Which will show you all of the parameters that you can view and change.

Parameters for model at 'my_awesome_model'
    model parameters:
    block parameters:
        Block: 'my_awesome_model/Constant' | Parameter: 'Value' | data_type: 'float64 (double) dims: [3, 3, 2] order: rtwCAPI_Orientation.col_major_nd'
        Block: 'my_awesome_model/Constant2' | Parameter: 'Value' | data_type: 'float64 (double) dims: [3, 4, 2] order: rtwCAPI_Orientation.col_major_nd'
        Block: 'my_awesome_model/Constant4' | Parameter: 'Value' | data_type: 'float64 (double) dims: [3, 4, 3] order: rtwCAPI_Orientation.col_major_nd'
    signals:
        Block: 'my_awesome_model/Clock' | Signal Name: '' | data_type: 'float64 (double) dims: [1, 1] order: rtwCAPI_Orientation.scalar'
        Block: 'my_awesome_model/Clock1' | Signal Name: '' | data_type: 'float64 (double) dims: [1, 1] order: rtwCAPI_Orientation.scalar'
    ...

Now you can run the model and start manipulating parameters.

from pysimlink import Model
import numpy as np

my_awesome_model = Model("my_awesome_model", "./my_awesome_model.zip")
my_awesome_model.reset()

for i in range(len(my_awesome_model)):
    Constant = my_awesome_model.get_block_param("my_awesome_model/Constant", param="Value")
    print(Constant) # np.ndarray
    new_val = np.full((3,3,2), i)
    my_awesome_model.set_block_param("my_awesome_model/Constant", param="Value", value=new_val)

Have a Generation error? Couldn’t find a compiler or nmake -? not found? See the compiler setup page.