rics.ml.time_split.types#
Types related to splitting data.
Module Attributes
Types that may be cast to |
|
Iterable that may be cast to |
|
Types that may be cast to |
|
User schedule type. |
|
User span type. |
|
Flexibility frequency string for |
|
A list of bounds. |
Classes
|
A 3-tuple which denotes two adjacent datetime ranges. |
|
Relative importance of data and future_data. |
- DatetimeTypes#
Types that may be cast to
pandas.Timestamp.
- DatetimeIterable#
Iterable that may be cast to
pandas.DatetimeIndex.alias of
Iterable[Union[str,Timestamp,datetime,date,datetime64]]
- TimedeltaTypes#
Types that may be cast to
pandas.Timedelta.alias of
Union[str,Timedelta,timedelta,timedelta64]
- Schedule#
User schedule type.
alias of
Union[DatetimeIndex,Iterable[Union[str,Timestamp,datetime,date,datetime64]],str,Timedelta,timedelta,timedelta64]
- Span#
User span type. Used to determine limits from the timestamps given by a
Schedule.alias of
Union[int,Literal[‘all’],str,Timedelta,timedelta,timedelta64]
- Flex#
Flexibility frequency string for
floor/ceil. PassFalseto disable.Options ‘auto’ (default) and
Trueare equivalent. Available data limits(start, end)are rounded usingTimestamp.floorandTimestamp.ceil, respectively. In other words, the start of the range is always rounded down while the end is rounded up.- Auto flex:
When using
flex='auto'(default), if and how bounds are expanded is determined by the size of the range and by distance between the actual and desired limits.Use
settings.auto_flexto modify auto-flex behaviour.- Manual flex:
Pass an offset alias specify how limits should be rounded. To specify by how much limits may be rounded, pass
'<'follow by a validpandas.Timedelta-string.See
expand_limits()for usage examples.
- class DatetimeSplitBounds(start: Timestamp, mid: Timestamp, end: Timestamp)[source]#
Bases:
NamedTupleA 3-tuple which denotes two adjacent datetime ranges.
- DatetimeSplits#
A list of bounds.
alias of
List[DatetimeSplitBounds]