Checks validity of input. Parameters ---------- x : `numpy.array`, `pandas.DataFrame` or `pandas.Series` The desi

Removes events that have too few datapoints or inconsistent / negligible scores given ``score_result`` and ``score_result_avg``. Para

Pickles a class when it can not be directly pickled by `dill`. Generates a dictionary with two keys. - {`obj_name`}.type stores t

Pickles a dictionary when it can not be directly pickled by `dill`. Generates a dictionary with two keys. - {`obj_name`}.type sto

Serializes a list or a tuple, preserving its order, when it can not be directly pickled by `dill`. Generates a dictionary with two keys.

Pickles an ordered dictionary when it can not be directly pickled by `dill`. Generates a dictionary with two keys. - {`obj_name`}

Trains the Multistage Forecast model with the given configurations. Parameters ---------- df : `pandas.DataFrame`

Validates model parameters. This function checks the following class attributes: - self.lag_unit - self.lags

This function adds the new anomaly columns for each metric. This will be applied to ``df`` once for each metric below. Parameters

Adds the covariance matrix for estimated coefficients to `info_dict` for linear models. The covariance matrix for estimated coefficients is defin

Adds features to one joined dataframe. This will be used to add features to all joined dataframes. Classes inherting from this class c

Adds the tests and confidence intervals for estimated coefficients to `info_dict` for linear models. Only tests and confidence intervals i

Adds coefficient summary df to `info_dict` for tree models. Only coefficient summary information will be added in this function. A series of

Adds degrees of freedom of the regression model to ``info_dict`` for linear models. The df is defined as the trace of hat matrix. In general, we

Adds model parameter information to ``info_dict`` for linear models. Only model-related information will be added in this function. A series

Adds model parameters to `info_dict` for tree models. Only model-related parameter information will be added in this function. A series of th

Adds model significance metrics to `info_dict` for linear models. Only model significance metrics information will be added in this function.

Adds model sum of squared errors to ``info_dict`` for linear models. Only sum of squared error information will be added in this function. A

This function will add a dummy column for the time periods of each group to the ``labels_df``. Parameters ---------- labels_df

This function does two things: - (1) adds shifted events to ``daily_event_df_dict`` and returns the new event dictionary. - (2) retur

Sets default values for ``model_components``. Parameters ---------- model_components : :class:`~greykite.framework.templates.

Applies the default ComputationParam values to the given object. If an expected attribute value is provided, the value is unchanged. Otherwise

Applies the default EvaluationMetricParam values to the given object. If an expected attribute value is provided, the value is unchanged. Othe

Applies the default EvaluationPeriodParam values to the given object. If an expected attribute value is provided, the value is unchanged. Othe

Applies the default MetadataParam values to the given object. If an expected attribute value is provided, the value is unchanged. Otherwise th

Sets default values for ``model_components``. Called by ``get_hyperparameter_grid`` after ``time_properties` is defined. Requires ``t

Asserts the forecast config values. This function expects a particular config and is not generic

Asserts the forecast config values. This function expects a particular config and is not generic

Asserts the forecast config values. This function expects a particular config and is not generic

Raises an AssertionError if the two forecast pipeline results are not equal

It assigns anomaly label to any points which has a larger mape than the threshold. Parameters ---------- data : See c

Prints out the holiday groups that contain holidays matching ``holiday_name_pattern`` and their scores. The searching is limited to the given

Spot checks the score of certain holidays containing pattern ``holiday_name_pattern``. Prints out the dates, individual day scores of all occu

Computes __evaluation_metrics for train and test set separately. :return: dictionary with train and test evaluation metrics

Computes the confusion matrix for two arrays. Parameters ---------- y_true : array-like, 1-D The actual categories. y_pred :

Creates the coefficient summary df section for model summary. Parameters ---------- info_dict : `dict` The dictionary returned by

Creates a information dictionary for linear model results. Only basic information will be created in this function. A series of these functio

Creates a information dictionary for tree model results. Only basic information will be created in this function. A series of these functions

Creates the model parameter section for model summary. Parameters ---------- info_dict : `dict` The dictionary returned by

Creates the residual section for model summary. Parameters ---------- info_dict : `dict` The dictionary returned by `~gre

Creates the model sifnificance section for model summary. Parameters ---------- info_dict : `dict` The dictionary returned by

Creates the title section for model summary. Returns ------- content : `str` Title section.

Creates the warning section for model summary. The following warnings are possible to be included: Condition number is too big.

Runs cross validation for one algo, data pair.

This is an abstract method to be implemented by children of this base class. Parameters ---------- y_new : `pandas.series`

The estimator instance to use as the final step in the pipeline. An instance of `greykite.sklearn.estimator.base_forecast_estimator.BaseForeca

Generate RST for a list of image filenames. Depending on whether we have one or more figures, we use a single rst call to 'image' or a horizon

Given a list of indices, with some of them being consecutive, find the start and end of each block. Indices are considered to be in the same b

Fits a model to training data Also fits the null model, if specified, for use in evaluating the `score` function Every subclass must

Fits the uncertainty model. Parameters ---------- train_df : `pandas.DataFrame` The training data.

Fits the uncertainty model. Parameters ---------- train_df : `pandas.DataFrame` The data used to fit the uncertai

Parameters ---------- data : `~greykite.detection.detector.ForecastDetectorData` Object including the data.

Fits a forecast and calculates cross validation errors for a given data set and ``fit_algorithm``. It returns a tuple: (Test MAPE, Fea

Fits a daily model for this use case. The daily model is a generic silverkite model with regressors.

This is an abstract method to be implemented by children of this base class. This logic is to be applied to `y_ready_to_fit` and it will updat

Returns the kth time period forecast. Parameters ---------- k : `int` The time period for which w

Generates a time series data frame by simulation and estimates quantiles by simulation and annotates the former with outliers :param freq:

Generates a Pandas DataFrame that represents time series data with arbitrary trends and level shifts. Example Usage: Calling `generate_df_with_arb

Gets the interaction terms between holidays and autoregression terms or other lag terms. Parameters ---------- daily_event_df_dict : `Dic

Returns a dictionary mapping each country to its holidays between the years specified. :param countries: List[str] countries for

Returns the indices of a node's children in the tree. Parameters ---------- node : `int` Index of the node.

Calculates comparison df with actuals and forecasted for the given horizon, using the training data up to time ``m`` ----------

Default parameter sets for benchmarking

Default parameter sets to framework.benchmark real datasets. The datasets are located in data folder. Every tuple has the following structure:

Default parameter sets for benchmarking silverkite template

Default parameter sets to generate simulated data for benchmarking. The training periods and forecast horizon are chosen to complement default rea

Gets the coefficients dataframe for Elastic Net regression models. Currently only returns the predictor names and their corresponding estimat

For each year, it returns the last Sunday in October, which is the end of the daylight saving (DST) in Europe. We assume Europe DST ends on l

For each year, it returns the last Sunday in March, which is the start of the daylight saving (DST) in Europe. We assume Europe DST starts on

Gets the coefficients dataframe for generalized linear models. The dataframe includes the estimated values, the standard errors, the Z-test v

Computes the score of all holiday events and their neighboring days in ``self.expanded_holiday_df``, by comparing their observed values with a

Returns hyperparameter grid. Implements the method in `~greykite.framework.templates.base_template.BaseTemplate`. Uses ``self.time_p

Returns hyperparameter grid. Implements the method in `~greykite.framework.templates.base_template.BaseTemplate`. Uses ``self.config

Returns hyperparameter grid. To be implemented by subclass. Available parameters: - self.df - self.config

Get the ``info_dict`` dictionary for tree models. A series of functions are used in a flow to get all information needed in tree model summar

Returns lagged regressor column names and minimal/maximal lag order. The lag order can be used to check potential imputation in the computatio

Gets the lagged regressor info for the model Iterates over each submodel to extract the lagged regressor info. Returns -----

Gets the lasso coefficients' and metrics on a single sample-splitting. Randomly splits the data into two parts of equal size. Performs the same

Gets the coefficients dataframe for lasso regression models. The dataframe includes the estimated values, the probability a coefficient is nonzer

Gets the lasso coefficients' p-values and confidence intervals by multi sample-splitting. Repeats the single sample-splitting in `~greykite.a

Returns a node's level in the tree. Level is defined as the length of the path to the root. The root is at level 0. Parameter

Gets the coefficients dataframe for least squared models. The dataframe includes the estimated values, the standard errors, the t-test values

Returns regressor column names from the model components. LagBasedTemplate does not support regressors.

Returns regressor column names. To be implemented by subclass. Available parameters: - self.df - self.confi

Gets the regressor columns in the model. Iterates over each submodel to extract the regressor columns. Returns -------

Gets the coefficients dataframe for ridge regression models. The dataframe includes the estimated values, the standard errors, the significan

Gets the ridge regression coefficients p-values and confidence intervals. The tests and confidence intervals are bootstrap p-values and confidenc

Return the SilverkiteColumn constants

Return the SilverkiteComponentsEnum constants

Return the SilverkiteHoliday constants

Return the SilverkiteHoliday constants