podpac.coordinates.StackedCoordinates¶

class podpac.coordinates.StackedCoordinates(coords, name=None, dims=None, ctype=None)[source]¶

Bases: podpac.core.coordinates.base_coordinates.BaseCoordinates

Stacked coordinates.

StackedCoordinates contain coordinates from two or more different dimensions that are stacked together to form a list of points (rather than a grid). The underlying coordinates values are Coordinates1d objects of equal size. The name for the stacked coordinates combines the underlying dimensions with underscores, e.g. 'lat_lon' or 'lat_lon_time'.

When creating Coordinates, podpac automatically detects StackedCoordinates. The following Coordinates contain 3 stacked lat-lon coordinates and 2 time coordinates in a 3 x 2 grid:

>>> lat = [0, 1, 2]
>>> lon = [10, 20, 30]
>>> time = ['2018-01-01', '2018-01-02']
>>> podpac.Coordinates([[lat, lon], time], dims=['lat_lon', 'time'])
Coordinates
    lat_lon[lat]: ArrayCoordinates1d(lat): Bounds[0.0, 2.0], N[3], ctype['midpoint']
    lat_lon[lon]: ArrayCoordinates1d(lon): Bounds[10.0, 30.0], N[3], ctype['midpoint']
    time: ArrayCoordinates1d(time): Bounds[2018-01-01, 2018-01-02], N[2], ctype['midpoint']

For convenience, you can also create uniformly-spaced stacked coordinates using clinspace:

>>> lat_lon = podpac.clinspace((0, 10), (2, 30), 3)
>>> time = ['2018-01-01', '2018-01-02']
>>> podpac.Coordinates([lat_lon, time], dims=['lat_lon', 'time'])
Coordinates
    lat_lon[lat]: ArrayCoordinates1d(lat): Bounds[0.0, 2.0], N[3], ctype['midpoint']
    lat_lon[lon]: ArrayCoordinates1d(lon): Bounds[10.0, 30.0], N[3], ctype['midpoint']
    time: ArrayCoordinates1d(time): Bounds[2018-01-01, 2018-01-02], N[2], ctype['midpoint']

Parameters

dims (tuple) – Tuple of dimension names.
name (str) – Stacked dimension name.
coords (dict-like) – xarray coordinates (container of coordinate arrays)
coordinates (pandas.MultiIndex) – MultiIndex of stacked coordinates values.

Alternative Constructors

`from_definition`(d)	Create StackedCoordinates from a stacked coordinates definition.
`from_xarray`(xcoords[, ctype])	Convert an xarray coord to StackedCoordinates

Methods

`__init__`(coords[, name, dims, ctype])	Initialize a multidimensional coords bject.
`copy`()	Make a copy of the stacked coordinates.
`select`(bounds[, return_indices, outer])	Get the coordinate values that are within the given bounds in all dimensions.
`transpose`(dims, *kwargs)	Transpose (re-order) the dimensions of the StackedCoordinates.

Attributes

`area_bounds`	Dictionary of (low, high) coordinates area_bounds in each dimension
`bounds`	Dictionary of (low, high) coordinates bounds in each dimension
`coordinates`	MultiIndex of stacked coordinates values.
`coords`	xarray coordinates (container of coordinate arrays)
`definition`	Serializable stacked coordinates definition.
`dims`	Tuple of dimension names.
`full_definition`	Serializable stacked coordinates definition, containing all properties.
`idims`	Tuple of indexing dimensions.
`name`	Stacked dimension name.
`shape`	Shape of the stacked coordinates.
`size`	Number of stacked coordinates.
`udims`	Tuple of unstacked dimension names, for compatibility.
`values`	MultiIndex of stacked coordinates values.

Members

__init__(coords, name=None, dims=None, ctype=None)[source]¶

Initialize a multidimensional coords bject.

Parameters

coords (list, StackedCoordinates) – Coordinate values in a list, or a StackedCoordinates object to copy.
ctype (str, optional) – Default coordinates type.