xarray_coarsened

PlasmaCalcs.tools.xarray_tools.xarray_dimensions.xarray_coarsened(array, dim, window_len, dim_coarse='window', dim_fine=None, *, assign_coarse_coords=False, boundary=UNSET, side=UNSET, stride=UNSET, fill_value=UNSET, keep_attrs=UNSET)

construct a coarsened version of array, where dim is coarsened by window_len,

and becomes two dims: dim_coarse and dim_fine.

Original dim coords will be associated with dim_coarse and dim_fine in the new array.

dim: str

dimension to coarsen.

if a non-dimension coordinate, will attempt to promote it to a dimension (e.g. via swap_dims).

window_len: int

length of the window to coarsen over.

dim_coarse: str, default ‘window’

name of coarse dimension; the i’th value here corresponds to the i’th window.

dim_fine: None or str

name of fine dimension; the j’th value here corresponds to the j’th element within a window.

if None, use ‘_’+dim, e.g. dim=’t’ –> dim_fine=’_t’.

assign_coarse_coords: bool or coords

coords to assign along the dim_coarse dimension.

True –> use np.arange.

False –> don’t assign coords.

boundary, side: UNSET or value

if provided (not UNSET), pass this value to coarsen().

boundary should be ‘exact’, ‘trim’, or ‘pad’.

side should be ‘left’ or ‘right’.

stride, fill_value, keep_attrs: UNSET or value

if provided (not UNSET), pass this value to construct().

docs for coarsen and construct are copied below, for convenience:

xarray.DataArray.coarsen

————————

Coarsen object for DataArrays.

Parameters

———-

dimmapping of hashable to int, optional

Mapping from the dimension name to the window size.

boundary{“exact”, “trim”, “pad”}, default: “exact”

If ‘exact’, a ValueError will be raised if dimension size is not a

multiple of the window size. If ‘trim’, the excess entries are

dropped. If ‘pad’, NA will be padded.

side : {“left”, “right”} or mapping of str to {“left”, “right”}, default: “left” coord_func : str or mapping of hashable to str, default: “mean”

function (name) that is applied to the coordinates,

or a mapping from coordinate name to function (name).

Returns

——-

core.rolling.DataArrayCoarsen

xr.core.rolling.DataArrayRolling.construct

——————————————

Convert this Coarsen object to a DataArray or Dataset,

where the coarsening dimension is split or reshaped to two

new dimensions.

Parameters

———-

window_dim: mapping

A mapping from existing dimension name to new dimension names.

The size of the second dimension will be the length of the

coarsening window.

keep_attrs: bool, optional

Preserve attributes if True

**window_dim_kwargs : {dim: new_name, …}

The keyword arguments form of `window_dim`.

Returns

——-

Dataset or DataArray with reshaped dimensions