pandas_extension.ResultFrameAggregator

pandas_extension.ResultFrameAggregator(
    self,
    agg=None,
    override_name=None,
    **kwargs,
)

Aggregates a MIKE IO 1D result DataFrame with hierarchical columns along specific levels.

Levels are categorized as entity levels, quantity levels, and aggregation levels. Aggregation is only performed along aggregation levels. Quantity levels define the resulting DataFrame’s columns. Entity levels define the resulting DataFrame’s indices.

Parameters

Name	Type	Description	Default
agg	str or callable	Default aggregation function to be applied along DataFrame column levels. Any str or callable accepted by pd.DataFrame.agg may be used. Example: - “max” : Take the maximum value. - “min” : Take the minimum value. - np.mean: Take the mean value. If not specified, then the ‘time’ parameter must be provided and is used as default.	None
override_name	str	Set a custom name for the overall aggregation. By default uses the agg function name.	None
**kwargs		Aggregation functions for specific DataFrame column levels (e.g. time=‘min’, chainage=‘mean’)	{}

Attributes

Name	Type	Description
entity_levels	list of str	The DataFrame column levels used to uniquely identify an entity. (e.g. [‘group’,‘name’,‘tag’]).
quantity_levels	list or str	The DataFrame column levels used to uniquely identify a quantity (e.g. [‘quantity’,‘derived’]).
agg_levels	list of str	The DataFrame column levels that will be aggregated along, in order. (e.g. [‘duplicate’,‘chainage’,‘time’]).
agg_functions	dict of str: callable	A dictionary with keys matching agg_levels, and values being the aggregation functions.

Examples

See which levels will be aggregated, and in what order.

>>> agg = ResultFrameAggregator('max')
>>> agg.agg_levels
['duplicate', 'chainage', 'time']

Aggregate along duplicate, chainage, and time, taking the max of each quantity

>>> agg = ResultFrameAggregator('max')
>>> agg.aggregate(df)

Always take the first chainage value, but take the max of all other levels.

>>> agg = ResultFrameAggregator('max', chainage='first')
>>> agg.aggregate(df)

Same result as above, but with explicit argument names.

>>> agg = ResultFrameAggregator(duplicate='max', time='max', chainage='first')
>>> agg.aggregate(df)

Same as above, but recognizing that time=‘max’ becomes the default for unspecifed levels.

>>> agg = ResultFrameAggregator(chainage='first' time='max')
>>> agg.aggregate(df)

Methods

Name	Description
aggregate	Aggregate along the duplicate, chainage, and time dimensions.
get_agg_function	Get the aggregation function for a level.
set_agg_function	Set the aggregation function for a level.

aggregate

pandas_extension.ResultFrameAggregator.aggregate(df)

Aggregate along the duplicate, chainage, and time dimensions.

get_agg_function

pandas_extension.ResultFrameAggregator.get_agg_function(level_name)

Get the aggregation function for a level.

Parameters

Name	Type	Description	Default
level_name	str	The level name to aggregate along. Must be one of the agg_levels.	required

Returns

Name	Type	Description
agg	pd.DataFrame.agg func-like	The aggregation function.

set_agg_function

pandas_extension.ResultFrameAggregator.set_agg_function(level_name, agg)

Set the aggregation function for a level.

Parameters

Name	Type	Description	Default
level_name	str	The level name to aggregate along. Must be one of the agg_levels.	required
agg	pd.DataFrame.agg func-like	The aggregation function.	required