deb-python-pint/docs/measurement.rst

Using Measurements

Measurements are the combination of two quantities: the mean value and the error (or uncertainty). The easiest ways to generate a measurement object is from a quantity using the plus_minus operator.

>>> import numpy as np >>> from pint import UnitRegistry >>> ureg = UnitRegistry() >>> book_length = (20. * ureg.centimeter).plus_minus(2.) >>> print(book_length) (20.0 +/- 2.0) centimeter

import numpy as np from pint import UnitRegistry ureg = UnitRegistry() Q = ureg.Quantity

You can inspect the mean value, the absolute error and the relative error:

>>> print(book_length.value) 20.0 centimeter >>> print(book_length.error) 2.0 centimeter >>> print(book_length.rel) 0.1

You can also create a Measurement object giving the relative error:

>>> book_length = (20. * ureg.centimeter).plus_minus(.1, relative=True) >>> print(book_length) (20.0 +/- 2.0) centimeter

Measurements support the same formatting codes as Quantity. For example, to pretty print a measurement with 2 decimal positions:

>>> print('{:.02fP}'.format(book_length)) (20.00 ± 2.00) centimeter

Mathematical operations with Measurements, return new measurements following the Propagation of uncertainty rules.

>>> print(2 * book_length) (40.0 +/- 4.0) centimeter >>> width = (10 * ureg.centimeter).plus_minus(1) >>> print('{:.02f}'.format(book_length + width)) (30.00 +/- 2.24) centimeter

Note

only linear combinations are currently supported.