How to define a piecewise anonymous function

74 visualizaciones (últimos 30 días)
Stephan
Stephan el 31 de Mayo de 2019
Comentada: Roy Goodman el 30 de Oct. de 2025 a las 2:45
Hello everyone,
the example code
syms x
continuous_function = x^2+x;
matlabFunction(continuous_function,'Vars',x)
gives me the anonymous function
@(x)x+x.^2
I would like a similar result for piecewise functions. However, the code
syms x
piecewise_function = piecewise( 0<x<1, x, 'otherwiseVal', 0 );
matlabFunction(piecewise_function,'Vars',x)
does not work.
Thanks for any help!

Respuesta aceptada

Star Strider
Star Strider el 31 de Mayo de 2019
You need to use ‘logical indexing’:
piecewise_function = @(x) (x.^2+x) .* ((0<x) & (x<1));
x = linspace(-1, 2);
figure
plot(x, piecewise_function(x))
producing:
How to define a piecewise anonymous function - 2019 05 31.png
  11 comentarios
Walter Roberson
Walter Roberson el 29 de Oct. de 2025 a las 22:15
syms x
f = piecewise((0<=x) & (x<=1),x.^2+x,nan)
f = 
piecewise_function = matlabFunction(f, 'file', 'test_piecewise.m')
piecewise_function = function_handle with value:
@test_piecewise
dbtype test_piecewise
1 function f = test_piecewise(x) 2 %TEST_PIECEWISE 3 % F = TEST_PIECEWISE(X) 4 5 % This function was generated by the Symbolic Math Toolbox version 25.2. 6 % 29-Oct-2025 22:14:42 7 8 t2 = ((0.0 <= x) & (x <= 1.0)); 9 if ~all(cellfun(@isscalar,{t2})) 10 error(message('symbolic:sym:matlabFunction:ConditionsMustBeScalar')); 11 end 12 if (t2) 13 f = x+x.^2; 14 else 15 f = NaN; 16 end 17 end
x = linspace(-1, 2);
plot(x, arrayfun(piecewise_function, x))
xline([-1 2])
Roy Goodman
Roy Goodman el 30 de Oct. de 2025 a las 2:45
Thanks!

Iniciar sesión para comentar.

Más respuestas (1)

Walter Roberson
Walter Roberson el 16 de Mayo de 2020
If you use matlabFunction with 'file' option, then it will convert piecewise() into if/else in the generated code.
Note: because it uses if/else instead of logical indexing, the generated code will not be vectorized on that variable.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by