pymc.math.erf#

pymc.math.erf = Elemwise(scalar_op=erf,inplace_pattern=<frozendict {}>)[source]#

error function

Computes the element-wise error function of a tensor.

Parameters:

aTensorLike: Input tensor

Returns:

TensorVariable: Output tensor with the error function evaluated at each element, with values in the range [-1, 1]

Notes

This function corresponds to SciPy’s scipy.special.erf function. The error function is defined as: erf(x) = (2/√π) * ∫(0 to x) exp(-t²) dt

Examples

>>> import pytensor
>>> import pytensor.tensor as pt
>>> x = pt.vector("x")
>>> f = pytensor.function([x], pt.erf(x))
>>> f([-1, 0, 1])
array([-0.84270079,  0.        ,  0.84270079])