Base Methods

class pycast.methods.basemethod.BaseMethod(requiredParameters=None, hasToBeSorted=True, hasToBeNormalized=True)

Bases: pycast.common.pycastobject.PyCastObject

Baseclass for all smoothing and forecasting methods.

__init__(requiredParameters=None, hasToBeSorted=True, hasToBeNormalized=True)

Initializes the BaseMethod.

Parameters:

• requiredParameters (list) – List of parameternames that have to be defined.
• hasToBeSorted (boolean) – Defines if the TimeSeries has to be sorted or not.
• hasToBeNormalized (boolean) – Defines if the TimeSeries has to be normalized or not.

_get_parameter_intervals()

Returns the intervals for the methods parameter.

Only parameters with defined intervals can be used for optimization!

Returns:Returns a dictionary containing the parameter intervals, using the parameter name as key, while the value hast the following format: [minValue, maxValue, minIntervalClosed, maxIntervalClosed]

• minValue

  Minimal value for the parameter

• maxValue

  Maximal value for the parameter

• minIntervalClosed

  True, if minValue represents a valid value for the parameter. False otherwise.

• maxIntervalClosed:

  True, if maxValue represents a valid value for the parameter. False otherwise.

Return type:dictionary

_get_value_error_message_for_invalid_prarameter(parameter, value)

Returns the ValueError message for the given parameter.

Parameters:

• parameter (string) – Name of the parameter the message has to be created for.
• value (numeric) – Value outside the parameters interval.

Returns:

Returns a string containing hte message.

Return type:

string

_in_valid_interval(parameter, value)

Returns if the parameter is within its valid interval.

Parameters:

• parameter (string) – Name of the parameter that has to be checked.
• value (numeric) – Value of the parameter.

Returns:

Returns True it the value for the given parameter is valid, False otherwise.

Return type:

boolean

can_be_executed()

Returns if the method can already be executed.

Returns:Returns True if all required parameters where already set, False otherwise. Return type:boolean

execute(timeSeries)

Executes the BaseMethod on a given TimeSeries object.

Parameters:timeSeries (TimeSeries) – TimeSeries object that fullfills all requirements (normalization, sortOrder). Returns:Returns a TimeSeries object containing the smoothed/forecasted values. Return type:TimeSeries Raise:Raises a NotImplementedError if the child class does not overwrite this function.

get_interval(parameter)

Returns the interval for a given parameter.

Parameters:parameter (string) – Name of the parameter. Returns:Returns a list containing with [minValue, maxValue, minIntervalClosed, maxIntervalClosed]. If no interval definitions for the given parameter exist, None is returned.

• minValue

  Minimal value for the parameter

• maxValue

  Maximal value for the parameter

• minIntervalClosed

  True, if minValue represents a valid value for the parameter. False otherwise.

• maxIntervalClosed:

  True, if maxValue represents a valid value for the parameter. False otherwise.

Return type:list

get_parameter(name)

Returns a forecasting parameter.

Parameters:name (string) – Name of the parameter. Returns:Returns the value stored in parameter. Return type:numeric Raise:Raises a KeyError if the parameter is not defined.

get_required_parameters()

Returns a list with the names of all required parameters.

Returns:Returns a list with the names of all required parameters. Return type:list

has_to_be_normalized()

Returns if the TimeSeries has to be normalized or not.

Returns:Returns True if the TimeSeries has to be normalized, False otherwise. Return type:boolean

has_to_be_sorted()

Returns if the TimeSeries has to be sorted or not.

Returns:Returns True if the TimeSeries has to be sorted, False otherwise. Return type:boolean

set_parameter(name, value)

Sets a parameter for the BaseMethod.

Parameters:

• name (string) – Name of the parameter that has to be checked.
• value (numeric) – Value of the parameter.

_interval_definitions = {False: ['(', ')'], True: ['[', ']']}

class pycast.methods.basemethod.BaseForecastingMethod(requiredParameters=None, valuesToForecast=1, hasToBeSorted=True, hasToBeNormalized=True)

Bases: pycast.methods.basemethod.BaseMethod

Basemethod for all forecasting methods.

__init__(requiredParameters=None, valuesToForecast=1, hasToBeSorted=True, hasToBeNormalized=True)

Initializes the BaseForecastingMethod.

Parameters:

• requiredParameters (list) – List of parameternames that have to be defined.
• valuesToForecast (integer) – Number of entries that will be forecasted. This can be changed by using forecast_until().
• hasToBeSorted (boolean) – Defines if the TimeSeries has to be sorted or not.
• hasToBeNormalized (boolean) – Defines if the TimeSeries has to be normalized or not.

Raise:

Raises a ValueError when valuesToForecast is smaller than zero.

_calculate_values_to_forecast(timeSeries)

Calculates the number of values, that need to be forecasted to match the goal set in forecast_until.

This sets the parameter “valuesToForecast” and should be called at the beginning of the BaseMethod.execute() implementation.

Parameters:timeSeries (TimeSeries) – Should be a sorted and normalized TimeSeries instance. Raise:Raises a ValueError if the TimeSeries is either not normalized or sorted.

forecast_until(timestamp, format=None)

Sets the forecasting goal (timestamp wise).

This function enables the automatic determination of valuesToForecast.

Parameters:

• timestamp – timestamp containing the end date of the forecast.
• format (string) – Format of the timestamp. This is used to convert the timestamp from UNIX epochs, if necessary. For valid examples take a look into the time.strptime() documentation.

get_optimizable_parameters()

Returns a list with optimizable parameters.

All required parameters of a forecasting method with defined intervals can be used for optimization.

Returns:Returns a list with optimizable parameter names. Return type:list Todo:Should we return all parameter names from the self._parameterIntervals instead?

set_parameter(name, value)

Sets a parameter for the BaseForecastingMethod.

Parameters:

• name (string) – Name of the parameter.
• value (numeric) – Value of the parameter.