ak.from_iter
Defined in awkward.operations.convert on line 829.
- ak.from_iter(iterable, highlevel=True, behavior=None, allow_record=True, initial=1024, resize=1.5)
- Parameters:
iterable (Python iterable) – Data to convert into an Awkward Array.
highlevel (bool) – If True, return an
ak.Array; otherwise, return a low-levelak.layout.Contentsubclass.behavior (None or dict) – Custom
ak.behaviorfor the output array, if high-level.allow_record (bool) – If True, the outermost element may be a record (returning
ak.Recordorak.layout.Recordtype, depending onhighlevel); if False, the outermost element must be an array.initial (int) – Initial size (in bytes) of buffers used by
ak.layout.ArrayBuilder(seeak.layout.ArrayBuilderOptions).resize (float) – Resize multiplier for buffers used by
ak.layout.ArrayBuilder(seeak.layout.ArrayBuilderOptions); should be strictly greater than 1.
Converts Python data into an Awkward Array.
Internally, this function uses ak.layout.ArrayBuilder (see the high-level
ak.ArrayBuilder documentation for a more complete description), so it
has the same flexibility and the same constraints. Any heterogeneous
and deeply nested Python data can be converted, but the output will never
have regular-typed array lengths.
The following Python types are supported.
bool, including
np.bool_: converted intoak.layout.NumpyArray.int, including
np.integer: converted intoak.layout.NumpyArray.float, including
np.floating: converted intoak.layout.NumpyArray.bytes: converted into
ak.layout.ListOffsetArraywith parameter"__array__"equal to"bytestring"(unencoded bytes).str: converted into
ak.layout.ListOffsetArraywith parameter"__array__"equal to"string"(UTF-8 encoded string).tuple: converted into
ak.layout.RecordArraywithout field names (i.e. homogeneously typed, uniform sized tuples).dict: converted into
ak.layout.RecordArraywith field names (i.e. homogeneously typed records with the same sets of fields).iterable, including np.ndarray: converted into
ak.layout.ListOffsetArray.
See also ak.to_list.