Skip to content

continuous_timeseries.discrete_to_continuous.linear#

Conversion of discrete to continuous data assuming linear interpolation

Functions:

Name Description
discrete_to_continuous_linear

Convert a discrete timeseries to a piecewise linear

discrete_to_continuous_linear #

discrete_to_continuous_linear(
    x: PINT_NUMPY_ARRAY, y: PINT_NUMPY_ARRAY, name: str
) -> TimeseriesContinuous

Convert a discrete timeseries to a piecewise linear

For details, see the module's docstring.

Parameters:

Name Type Description Default
x PINT_NUMPY_ARRAY

The discrete x-values from which to convert

 required
y PINT_NUMPY_ARRAY

The discrete y-values from which to convert

 required
name str

The value to use to set the result's name attribute

 required

Returns:

Type Description
TimeseriesContinuous

Continuous version of discrete based on a piecewise linear interpolation.
Source code in src/continuous_timeseries/discrete_to_continuous/linear.py 
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
def discrete_to_continuous_linear(
    x: PINT_NUMPY_ARRAY,
    y: PINT_NUMPY_ARRAY,
    name: str,
) -> TimeseriesContinuous:
    """
    Convert a discrete timeseries to a piecewise linear

    For details, see
    [the module's docstring][continuous_timeseries.discrete_to_continuous.linear].

    Parameters
    ----------
    x
        The discrete x-values from which to convert

    y
        The discrete y-values from which to convert

    name
        The value to use to set the result's name attribute

    Returns
    -------
    :
        Continuous version of `discrete`
        based on a piecewise linear interpolation.
    """
    try:
        import scipy.interpolate
    except ImportError as exc:
        raise MissingOptionalDependencyError(
            "discrete_to_continuous_linear", requirement="scipy"
        ) from exc

    # Late import to avoid circularity
    from continuous_timeseries.timeseries_continuous import (
        ContinuousFunctionScipyPPoly,
        TimeseriesContinuous,
    )

    x_m = x.m
    time_steps = x[1:] - x[:-1]

    coeffs = np.zeros((2, y.size - 1))

    rises = y[1:] - y[:-1]

    coeffs[0, :] = (rises / time_steps).m
    coeffs[1, :] = y.m[:-1]

    piecewise_polynomial = scipy.interpolate.PPoly(
        x=x_m,
        c=coeffs,
        extrapolate=False,  # Avoid extrapolation by default
    )

    res = TimeseriesContinuous(
        name=name,
        time_units=x.u,
        values_units=y.u,
        function=ContinuousFunctionScipyPPoly(piecewise_polynomial),
        domain=(np.min(x), np.max(x)),
    )

    return res