constant_stf

Graph.constant_stf(constant, batch_dimension_count=0)

Create a constant sampleable tensor-valued function of time.

Parameters:

  • constant (number or np.ndarray or Tensor) – The constant value \(c\) of the function. To create a batch of \(B_1 \times \ldots \times B_n\) constant functions of shape \(D_1 \times \ldots \times D_m\), provide this constant parameter as an object of shape \(B_1\times\ldots\times B_n\times D_1\times\ldots\times D_m\).

  • batch_dimension_count (int, optional) – The number of batch dimensions, \(n\), in constant. If provided, the first \(n\) dimensions of constant are considered batch dimensions. Defaults to 0, which corresponds to no batch.

Returns:

An Stf representing the constant function \(f(t) = c\) for all time \(t\) (or a batch of functions, if you provide batch_dimension_count).

Return type:

Stf

See also

constant_pwc

Corresponding operation for Pwcs.

constant_stf_operator

Create a constant sampleable function from operators.

identity_stf

Create an Stf representing the identity function.

Notes

For more information on Stf nodes see the Working with time-dependent functions in Boulder Opal topic.