earthkit.data.xr_engine.builder¶

Attributes¶

`BACKEND_AND_XR_OPEN_DS_KWARGS`
`NON_XR_OPEN_DS_KWARGS`

Classes¶

`BackendDataBuilder`
`DatasetBuilder`
`MemoryBackendDataBuilder`	Build a dataset using memory backend.
`SingleDatasetBuilder`
`SplitDatasetBuilder`
`TensorBackendArray`	Mixin class that extends a class that defines a `shape` property to
`TensorBackendDataBuilder`	Build a dataset using memory backend.
`VariableBuilder`

Functions¶

from_earthkit(ds[, backend_kwargs, other_kwargs])

Create an xarray dataset from an earthkit fieldlist.

Module Contents¶

earthkit.data.xr_engine.builder.BACKEND_AND_XR_OPEN_DS_KWARGS = ['decode_times', 'decode_timedelta', 'drop_variables']¶

class earthkit.data.xr_engine.builder.BackendDataBuilder(ds, profile, dims, grid=None, fixed_local_attrs=None)¶

array_namespace¶

build()¶

abstractmethod build_values(*args, **kwargs)¶

check_tensor_coords(var_name, coord_name, tensor_coords)¶

collect_aux_coords()¶

coords()¶

dims¶

ds¶

dtype¶

field_coords¶

fixed_local_attrs¶

flatten_values¶

grid = None¶

pre_build_variable(ds_var, dims, name)¶

pre_build_variables()¶: Generate a builder for each variable.

prepare_tensor(ds, dims, name)¶

profile¶

tensor_coords¶

class earthkit.data.xr_engine.builder.DatasetBuilder(ds, profile, backend_kwargs=None)¶

direct_backend¶

ds¶

grid(ds)¶

grids¶

parse(ds, profile=None, full=False)¶

profile_kwargs¶

split_dims¶

class earthkit.data.xr_engine.builder.MemoryBackendDataBuilder(ds, profile, dims, grid=None, fixed_local_attrs=None)¶

Bases: BackendDataBuilder

Build a dataset using memory backend.

Each variable contain data values in memory copied from the input fields. As soon as a field’s values are copied, the field is immediately released and its memory is freed if the field supports this operation. Fields released in this way are not available for further access.

array_namespace¶

build()¶

build_values(tensor, var_dims, name)¶: Generate the data object stored in the xarray variable.

check_tensor_coords(var_name, coord_name, tensor_coords)¶

collect_aux_coords()¶

coords()¶

dims¶

ds¶

dtype¶

field_coords¶

fixed_local_attrs¶

flatten_values¶

grid = None¶

pre_build_variable(ds_var, dims, name)¶

pre_build_variables()¶: Generate a builder for each variable.

prepare_tensor(ds, dims, name)¶

profile¶

tensor_coords¶

earthkit.data.xr_engine.builder.NON_XR_OPEN_DS_KWARGS = ['split_dims', 'direct_backend']¶

class earthkit.data.xr_engine.builder.SingleDatasetBuilder(*args, from_xr=False, **kwargs)¶

Bases: DatasetBuilder

build()¶

direct_backend¶

ds¶

grid(ds)¶

grids¶

parse(ds, profile=None, full=False)¶

profile_kwargs¶

split_dims¶

class earthkit.data.xr_engine.builder.SplitDatasetBuilder(*args, backend_kwargs=None, other_kwargs=None)¶

Bases: DatasetBuilder

build()¶

direct_backend¶

ds¶

grid(ds)¶

grids¶

parse(ds, profile=None, full=False)¶

prepare(keys)¶

profile_kwargs¶

split()¶

split_dims¶

class earthkit.data.xr_engine.builder.TensorBackendArray(tensor, dims, shape, array_namespace, dtype, var_name)¶

Bases: xarray.backends.common.BackendArray

Mixin class that extends a class that defines a shape property to one that also defines ndim, size and __len__.

array_namespace¶

async async_get_duck_array(dtype=None)¶

abstractmethod async_getitem(key)¶

Async:

dims¶

dtype¶

get_duck_array(dtype=None)¶

lock¶

property nbytes¶

property ndim: int¶: Number of array dimensions.

See also

numpy.ndarray.ndim

shape¶

property size: int¶

Number of elements in the array.

Equal to np.prod(a.shape), i.e., the product of the array’s dimensions.