Equation

Suppose you have a variable named equation (which is an instance of Equation) that represents the following information in an equipment register for equipment that measures relative humidity. The corrected value depends on two variables (r and t) and the standard uncertainty is a constant.

<equation>
  <value variables="r t">r-0.71-0.04*r+3.4e-4*pow(r,2)+2.4e-3*t+1.3e-3*r*t</value>
  <uncertainty variables="">0.355</uncertainty>
  <unit>%rh</unit>
  <ranges>
    <range variable="r">
      <minimum>30</minimum>
      <maximum>80</maximum>
    </range>
    <range variable="t">
      <minimum>15</minimum>
      <maximum>25</maximum>
    </range>
  </ranges>
</equation>

You can access the unit, degrees of freedom and comment as attributes of equation

>>> equation.unit
'%rh'
>>> equation.degree_freedom
inf
>>> equation.comment
''

To evaluate an equation, call the appropriate attribute with the variable(s) that are required to evaluate the equation with

>>> equation.value.variables
('r', 't')
>>> equation.uncertainty.variables
()
>>> assert equation.value(r=50.3, t=20.4) == 49.8211466
>>> assert equation.uncertainty() == 0.355

A variable can have multiple values. Any sequence of numbers, i.e., a list, tuple, ndarray, etc., may be used (tip: using ndarray will improve performance since a copy of the values is not required),

>>> equation.value(r=[50.3, 52.1, 48.7], t=[20.4, 19.7, 20.0])
array([49.8211466, 51.6104604, 48.1625746])

the values of the variables do not need to be 1-dimensional arrays,

>>> equation.value(r=[(50.3, 52.1), (48.7, 47.9)], t=[(20.4, 19.7), (20.0, 19.6)])
array([[49.8211466, 51.6104604],
       [48.1625746, 47.3216314]])

and the array broadcasting rules of numpy also apply, i.e., multiple r values and a single t value

>>> equation.value(r=(50.3, 52.1, 48.7), t=20.4)
array([49.8211466, 51.6595514, 48.1888586])

If you forget to specify a variable (in the following case, t) a NameError will be raised,

>>> equation.value(r=50.3)
Traceback (most recent call last):
...
NameError: name 't' is not defined

however, if you specify more variables than are required to evaluate the equation, the additional variables are ignored

>>> equation.uncertainty(r=50.3, t=20.4)
array(0.355)

Notice in the last returned value that the result was printed as array(0.355) even though a single r and t value was specified (although these variables were ignored in this particular example, since the standard uncertainty is a constant, the principle remains the same if they were not ignored). All evaluated returned types are an instance of a numpy ndarray even if a single value is specified. These particular returned array instances are referred to as 0-dimensional array scalars in numpy terminology.

When evaluating an equation, the value(s) of the input variables are checked to ensure that the value(s) are within the ranges that the equation is valid for. The XML data above shows that the temperature, t, value must be in the range 15 to 25. If you evaluate the corrected value at t=30 a ValueError is raised

>>> equation.value.ranges
{'r': Range(minimum=30, maximum=80), 't': Range(minimum=15, maximum=25)}
>>> equation.value(r=50.3, t=30)
Traceback (most recent call last):
...
ValueError: The value 30.0 is not within the range [15, 25]

You can bypass range checking by including a check_range=False keyword argument

>>> equation.value(r=50.3, t=30, check_range=False)
array(50.4719306)

Equation dataclass

Equation(
    value: Evaluable,
    uncertainty: Evaluable,
    unit: str,
    degree_freedom: float = float("inf"),
    comment: str = "",
)

Represents the equation element in an equipment register.

Parameters:

Name	Type	Description	Default
value	Evaluable	The equation to evaluate to calculate the corrected value.	required
uncertainty	Evaluable	The equation to evaluate to calculate the standard uncertainty.	required
unit	str	The unit of the measured quantity.	required
degree_freedom	float	The degrees of freedom.	float('inf')
comment	str	A comment to associate with the equation.	''

comment class-attribute instance-attribute

comment: str = ''

A comment associated with the equation.

degree_freedom class-attribute instance-attribute

degree_freedom: float = float('inf')

The degrees of freedom.

uncertainty instance-attribute

uncertainty: Evaluable

The equation to evaluate to calculate the standard uncertainty.

unit instance-attribute

unit: str

The unit of the measured quantity.

value instance-attribute

value: Evaluable

The equation to evaluate to calculate the corrected value.

from_xml classmethod

from_xml(element: Element[str]) -> Equation

Convert an XML element into an Equation instance.

Parameters:

Name	Type	Description	Default
element	Element[str]	An equation XML element from an equipment register.	required

Returns:

Type	Description
Equation	The Equation instance.

Source code in src/msl/equipment/schema.py

@classmethod
def from_xml(cls, element: Element[str]) -> Equation:
    """Convert an XML element into an [Equation][msl.equipment.schema.Equation] instance.

    Args:
        element: An [equation][type_equation] XML element from an equipment register.

    Returns:
        The [Equation][msl.equipment.schema.Equation] instance.
    """
    # Schema forces order
    value = element[0]
    uncertainty = element[1]
    ranges = {
        r.attrib["variable"]: Range(minimum=float(r[0].text or -np.inf), maximum=float(r[1].text or np.inf))
        for r in element[3]
    }

    return cls(
        value=Evaluable(
            equation=value.text or "", variables=tuple(value.attrib["variables"].split()), ranges=ranges
        ),
        uncertainty=Evaluable(
            equation=uncertainty.text or "", variables=tuple(uncertainty.attrib["variables"].split()), ranges=ranges
        ),
        unit=element[2].text or "",
        degree_freedom=float(element[4].text or np.inf) if len(element) > 4 else np.inf,  # noqa: PLR2004
        comment=element.attrib.get("comment", ""),
    )

to_xml

to_xml() -> Element[str]

Convert the Equation class into an XML element.

Returns:

Type	Description
Element[str]	The Equation as an XML element.

Source code in src/msl/equipment/schema.py

def to_xml(self) -> Element[str]:
    """Convert the [Equation][msl.equipment.schema.Equation] class into an XML element.

    Returns:
        The [Equation][msl.equipment.schema.Equation] as an XML element.
    """
    attrib = {"comment": self.comment} if self.comment else {}
    e = Element("equation", attrib=attrib)
    value = SubElement(e, "value", attrib={"variables": " ".join(self.value.variables)})
    value.text = self.value.equation
    uncertainty = SubElement(e, "uncertainty", attrib={"variables": " ".join(self.uncertainty.variables)})
    uncertainty.text = self.uncertainty.equation
    unit = SubElement(e, "unit")
    unit.text = self.unit

    ranges = SubElement(e, "ranges")
    for name, _range in self.value.ranges.items():  # self.value.ranges and self.uncertainty.ranges are the same
        rng = SubElement(ranges, "range", attrib={"variable": name})
        mn = SubElement(rng, "minimum")
        mn.text = str(_range.minimum)
        mx = SubElement(rng, "maximum")
        mx.text = str(_range.maximum)

    if not isinf(self.degree_freedom):
        dof = SubElement(e, "degreeFreedom")
        dof.text = str(self.degree_freedom)

    return e

Evaluable dataclass

Evaluable(
    equation: str,
    variables: tuple[str, ...] = (),
    ranges: dict[str, Range] = dict(),
)

Represents the <value> and <uncertainty> XML elements in an equation.

Parameters:

Name	Type	Description	Default
equation	str	The string representation of the equation to evaluate.	required
variables	tuple[str, ...]	The names of the variables in the equation.	()
ranges	dict[str, Range]	The numeric range for a variable that the equation is valid for. The keys are the variable names. A range does not need to be defined for every variable. If a range is not defined then a range of \([-\infty, +\infty]\) is assumed.	dict()

equation instance-attribute

equation: str

The string representation of the equation to evaluate.

ranges class-attribute instance-attribute

ranges: dict[str, Range] = field(default_factory=dict)

The numeric range for each variable that the equation is valid for. The keys are the variable names.

variables class-attribute instance-attribute

variables: tuple[str, ...] = ()

The names of the variables in the equation.

Range

Bases: NamedTuple

The numeric range of a variable that an equation is valid for.

Parameters:

Name	Type	Description
minimum	float	Minimum value in range.
maximum	float	Maximum value in range.

maximum instance-attribute

maximum: float

Maximum value in range.

minimum instance-attribute

minimum: float

Minimum value in range.

check_within_range

check_within_range(
    value: float | ArrayLike,
) -> Literal[True]

Check that the values are within the range.

Parameters:

Name	Type	Description	Default
value	float | ArrayLike	The values to check, raises	required

Returns:

Type	Description
Literal[True]	Always returns True. Raises ValueError if value is not within the range.

Source code in src/msl/equipment/schema.py

def check_within_range(self, value: float | ArrayLike) -> Literal[True]:
    """Check that the values are within the range.

    Args:
        value: The values to check, raises

    Returns:
        Always returns `True`. Raises [ValueError][] if
            `value` is not within the range.
    """
    if isinstance(value, (int, float)) or (isinstance(value, np.ndarray) and value.ndim == 0):
        if value < self.minimum or value > self.maximum:
            msg = f"The value {value} is not within the range [{self.minimum}, {self.maximum}]"
            raise ValueError(msg)
    elif np.any(np.less(value, self.minimum)) or np.any(np.greater(value, self.maximum)):  # pyright: ignore[reportUnknownArgumentType]
        msg = f"A value in the sequence is not within the range [{self.minimum}, {self.maximum}]"
        raise ValueError(msg)
    return True