Make numpy ufunc work with ufloat #344
Description
-
Versions:
python == 3.10.12
numpy == 2.2.6
uncertainties == 3.2.3 -
Code to reproduce / observed behavior:
In [1]: import numpy
In [2]: from uncertainties import ufloat
In [3]: numpy.exp(ufloat(1., 0.1))
---------------------------------------------------------------------------
AttributeError Traceback (most recent call last)
AttributeError: 'Variable' object has no attribute 'exp'
The above exception was the direct cause of the following exception:
TypeError Traceback (most recent call last)
Cell In[3], line 1
----> 1 numpy.exp(ufloat(1., 0.1))
TypeError: loop of ufunc does not support argument 0 of type Variable which has no callable exp method- Expected behavior: Calling an ufunc from numpy should have similar behavior as calling the same function from
uncertainties.unumpy:
In [4]: from uncertainties import unumpy
In [5]: unumpy.exp(ufloat(1., 0.1))
Out[5]: array(2.718281828459045+/-0.27182818284590454, dtype=object)-
Motivation: using uncertainties with other packages (for example pint) will be facilitated enormously, for example to call
numpy.exp(pint.Quantity(ufloat(1., 0.1)))using this basic example. Of course, it would be nice if all numpy ufuncs and numpy functions work in this way. Currently, I have to maintain some custom fragile code to use numpy ufuncs and numpy functions with the two packages. -
Suggested solutions: one possible way is maybe to define some special methods in the
uncertainties.core.Variableclass to point to the custom unumpy implementations. This seems to work in pint.Quantity.