pandas.arrays.SparseArray#

classpandas.arrays.SparseArray(data,sparse_index=None,fill_value=None,kind='integer',dtype=None,copy=False)[source]#

An ExtensionArray for storing sparse data.

Parameters:

dataarray-like or scalar

A dense array of values to store in the SparseArray. This may containfill_value.

sparse_indexSparseIndex, optional

fill_valuescalar, optional

Elements in data that arefill_value are not stored in theSparseArray. For memory savings, this should be the most common valueindata. By default,fill_value depends on the dtype ofdata:

data.dtype	na_value
float	np.nan
int	0
bool	False
datetime64	pd.NaT
timedelta64	pd.NaT

The fill value is potentially specified in three ways. In order ofprecedence, these are

1. Thefill_value argument

2. dtype.fill_value iffill_value is None anddtype isaSparseDtype

3. data.dtype.fill_value iffill_value is None anddtypeis not aSparseDtype anddata is aSparseArray.

kindstr

Can be ‘integer’ or ‘block’, default is ‘integer’.The type of storage for sparse locations.

• ‘block’: Stores ablock andblock_length for eachcontiguousspan of sparse values. This is best whensparse data tends to be clumped together, with largeregions offill-value values between sparse values.

• ‘integer’: uses an integer to store the location ofeach sparse value.

dtypenp.dtype or SparseDtype, optional

The dtype to use for the SparseArray. For numpy dtypes, thisdetermines the dtype ofself.sp_values. For SparseDtype,this determinesself.sp_values andself.fill_value.

copybool, default False

Whether to explicitly copy the incomingdata array.

Attributes

None

Methods

None

Examples

>>>frompandas.arraysimportSparseArray>>>arr=SparseArray([0,0,1,2])>>>arr[0, 0, 1, 2]Fill: 0IntIndexIndices: array([2, 3], dtype=int32)