cudf.core.dtypes.Decimal64Dtype#

class cudf.core.dtypes.Decimal64Dtype(precision, scale=0)[source]#

Type to represent a decimal64 data.

Parameters:

precisionint: The total number of digits in each value of this dtype
scaleint, optional: The scale of the dtype. See Notes below.

Attributes

`precision`	The decimal precision, in number of decimal digits (an integer).
`scale`	The decimal scale (an integer).
`itemsize`	Length of one column element in bytes.

Methods

`to_arrow`()	Return the equivalent `pyarrow` dtype.
`from_arrow`(typ)	Construct a cudf decimal dtype from a `pyarrow` dtype

Notes

When the scale is positive:

numbers with fractional parts (e.g., 0.0042) can be represented
the scale is the total number of digits to the right of the decimal point

When the scale is negative:

only multiples of powers of 10 (including 10**0) can be represented (e.g., 1729, 4200, 1000000)
the scale represents the number of trailing zeros in the value.

For example, 42 is representable with precision=2 and scale=0. 13.0051 is representable with precision=6 and scale=4, and not representable with precision<6 or scale<4.

Examples

>>> import cudf
>>> decimal64_dtype = cudf.Decimal64Dtype(precision=9, scale=2)
>>> decimal64_dtype
Decimal64Dtype(precision=9, scale=2)