gemseo.disciplines.linear_combination module

Discipline computing a linear combination of its inputs.

class LinearCombination(input_names, output_name, input_coefficients=None, offset=0.0, input_size=None)

Bases: Discipline

Discipline computing a linear combination of its inputs.

The user can specify the coefficients related to the variables as well as the offset.

E.g., a discipline computing the output \(y\) from \(d\) inputs \(x_1,\ldots,x_d\) with the function \(f(x_1,\ldots,x_d)=a_0+\sum_{i=1}^d a_i x_i\).

When the offset \(a_0\) is equal to 0 and the coefficients \(a_1,\ldots,a_d\) are equal to 1, the discipline simply sums the inputs.

Notes

By default, the LinearCombination simply sums the inputs.

Examples

>>> discipline = LinearCombination(["alpha", "beta", "gamma"], "delta",
        input_coefficients={"alpha": 1.,"beta": 2.,"gamma": 3.})
>>> input_data = {"alpha": array([1.0]), "beta": array([1.0]),
        "gamma": array([1.0])}
>>> discipline.execute(input_data)
>>> delta = discipline.io.data["delta"]  # delta = array([6.])

Initialize self. See help(type(self)) for accurate signature.

Parameters:

• input_names (Iterable[str]) -- The names of input variables.

• output_name (str) -- The name of the output variable.

• input_coefficients (dict[str, float] | None) -- The coefficients related to the input variables. If None, use 1 for all the input variables.

• offset (float) --

  The output value when all the input variables are equal to zero.

  By default it is set to 0.0.

• input_size (int | None) -- The size of the inputs. If None, the default inputs are initialized with size 1 arrays.