A single character string for the type of model. The only possible value for this model is "regression".

committees

neighbors

max rules

external regressors

forecast horizon

selected features

A dataframe of xreg (exogenous regressors)

A numeric vector of values to fit

committees

neighbors

max rules

lag periods

external regressors

forecast horizon

selected features

model object

input data to predict

Additional parsnip-related options, depending on the value of type. Arguments to the underlying model's prediction function cannot be passed here (use the opts argument instead). Possible arguments are:

• interval: for type equal to "survival" or "quantile", should interval estimates be added, if available? Options are "none" and "confidence".

• level: for type equal to "conf_int", "pred_int", or "survival", this is the parameter for the tail area of the intervals (e.g. confidence level for confidence intervals). Default value is 0.95.

• std_error: for type equal to "conf_int" or "pred_int", add the standard error of fit or prediction (on the scale of the linear predictors). Default value is FALSE.

• quantile: for type equal to quantile, the quantiles of the distribution. Default is (1:9)/10.

• eval_time: for type equal to "survival" or "hazard", the time points at which the survival probability or hazard is estimated.

run info using the set_run_info() function

Default of NULL runs no parallel processing and forecasts each individual time series one after another. 'local_machine' leverages all cores on current machine Finn is running on. 'spark' runs time series in parallel on a spark cluster in Azure Databricks or Azure Synapse.

Run components of forecast process inside a specific time series in parallel. Can only be used if parallel_processing is set to NULL or 'spark'.

Number of cores to run when parallel processing is set up. Used when running parallel computations on local machine or within Azure. Default of NULL uses total amount of cores on machine minus one. Can't be greater than number of cores on machine minus 1.

Set seed for random number generator. Numeric value.

run info using the set_run_info() function.

If TRUE, create simple averages of individual models and save the most accurate one.

Max number of models to average together. Will create model averages for 2 models up until input value or max number of models ran.

If TRUE, convert a week forecast down to day by evenly splitting across each day of week. Helps when aggregating up to higher temporal levels like month or quarter.

Default of NULL runs no parallel processing and forecasts each individual time series one after another. 'local_machine' leverages all cores on current machine Finn is running on. 'spark' runs time series in parallel on a spark cluster in Azure Databricks or Azure Synapse.

Run components of forecast process inside a specific time series in parallel. Can only be used if parallel_processing is set to NULL or 'spark'.

Number of cores to run when parallel processing is set up. Used when running parallel computations on local machine or within Azure. Default of NULL uses total amount of cores on machine minus one. Can't be greater than number of cores on machine minus 1.

Run info using set_run_info()

A data frame or tibble of historical time series data. Can also include external regressors for both historical and future data.

List of column headers within input data to be used to separate individual time series.

The column header formatted as a character value within input data you want to forecast.

The date granularity of the input data. Finn accepts the following as a character string day, week, month, quarter, year.

Number of periods to forecast into the future.

List of column headers within input data to be used as features in multivariate models.

Date value of when your input_data starts. Default of NULL is to use earliest date value in input_data.

Date value of when your input_data ends.Default of NULL is to use the latest date value in input_data.

Date value to remove individual time series that don't contain non-zero values after that specified date. Default of NULL is to not remove any time series and attempt to forecast all of them.

Month number of start of fiscal year of input data, aids in building out date features. Formatted as a numeric value. Default of 1 assumes fiscal year starts in January.

If TRUE, cleans missing values. Only impute values for missing data within an existing series, and does not add new values onto the beginning or end, but does provide a value of 0 for said values. Turned off when running hierarchical forecasts.

If TRUE, outliers are cleaned and inputted with values more in line with historical data

Number of specific back test folds to run when determining the best model. Default of NULL will automatically choose the number of back tests to run based on historical data size, which tries to always use a minimum of 80% of the data when training a model.

Number of periods to move back for each back test scenario. Default of NULL moves back 1 period at a time for year, quarter, and month data. Moves back 4 for week and 7 for day data.

How Finn should approach your data. Current default and only option is 'accuracy'. In the future this could evolve to other areas like optimizing for interpretability over accuracy.

How the forecast is created. The default of 'bottoms_up' trains models for each individual time series. 'grouped_hierarchy' creates a grouped time series to forecast at while 'standard_hierarchy' creates a more traditional hierarchical time series to forecast, both based on the hts package.

Default of NULL runs no parallel processing and forecasts each individual time series one after another. 'local_machine' leverages all cores on current machine Finn is running on. 'spark' runs time series in parallel on a spark cluster in Azure Databricks or Azure Synapse.

Run components of forecast process inside a specific time series in parallel. Can only be used if parallel_processing is set to NULL or 'spark'.

Number of cores to run when parallel processing is set up. Used when running parallel computations on local machine or within Azure. Default of NULL uses total amount of cores on machine minus one. Can't be greater than number of cores on machine minus 1.

If TRUE, log transform target variable before training models.

If TRUE, allow forecasts to dip below zero.

List of values to use in creating fourier series as features. Default of NULL automatically chooses these values based on the date_type.

List of values to use in creating lag features. Default of NULL automatically chooses these values based on date_type.

List of values to use in creating rolling window features. Default of NULL automatically chooses these values based on date type.

List of recipes to run on multivariate models that can run different recipes. A value of NULL runs all recipes, but only runs the R1 recipe for weekly and daily date types, and also for global models to prevent memory issues. A value of "all" runs all recipes, regardless of date type or if it's a local/global model. A list like c("R1") or c("R2") would only run models with the R1 or R2 recipe.

If TRUE, run principle component analysis on any lagged features to speed up model run time. Default of NULL runs PCA on day and week date types across all local multivariate models, and also for global models across all date types.

List of models to run. Default of NULL runs all models.

List of models not to run, overrides values in models_to_run. Default of NULL doesn't turn off any model.

If TRUE, run multivariate models on the entire data set (across all time series) as a global model. Can be override by models_not_to_run. Default of NULL runs global models for all date types except week and day.

If TRUE, run models by individual time series as local models.

If TRUE, run ensemble models. Default of NULL runs ensemble models only for quarter and month date types.

If TRUE, create simple averages of individual models.

Max number of models to average together. Will create model averages for 2 models up until input value or max number of models ran.

Implement feature selection before model training

If TRUE, convert a week forecast down to day by evenly splitting across each day of week. Helps when aggregating up to higher temporal levels like month or quarter.

Set seed for random number generator. Numeric value.

If TRUE, runs model training in parallel, only works when parallel_processing is set to 'local_machine' or 'spark'. Recommended to use a value of FALSE and leverage inner_parallel for new features.

If TRUE, return the forecast results. Used to be backwards compatible with previous finnts versions. Recommended to use a value of FALSE and leverage get_forecast_data() for new features.

Name used when submitting jobs to external compute like Azure Batch. Formatted as a character string.

run info using the set_run_info() function

return type

run info using the set_run_info() function

recipe to return. Either a value of "R1" or "R2"

return type

run info using the set_run_info() function

Name used to group similar runs under a single experiment name.

Name to distinguish one run of Finn from another. The current time in UTC is appended to the run name to ensure a unique run name is created.

Used to store outputs during a run to other storage services in Azure. Could be a storage container object from the 'AzureStor' package to connect to ADLS blob storage or a OneDrive/SharePoint object from the 'Microsoft365R' package to connect to a OneDrive folder or SharePoint site. Default of NULL will save outputs to the local file system.

String showing what file path the outputs should be written to. Default of NULL will write the outputs to a temporary directory within R, which will delete itself after the R session closes.

run info using the set_run_info() function

A single character string for the type of model. The only possible value for this model is "regression".

mixture

penalty

lag periods

external regressors

forecast horizon

selected features

A dataframe of xreg (exogenous regressors)

A numeric vector of values to fit

alpha

lambda

lag periods

external regressors

forecast horizon

selected features

model object

input data to predict

Additional parsnip-related options, depending on the value of type. Arguments to the underlying model's prediction function cannot be passed here (use the opts argument instead). Possible arguments are:

• interval: for type equal to "survival" or "quantile", should interval estimates be added, if available? Options are "none" and "confidence".

• level: for type equal to "conf_int", "pred_int", or "survival", this is the parameter for the tail area of the intervals (e.g. confidence level for confidence intervals). Default value is 0.95.

• std_error: for type equal to "conf_int" or "pred_int", add the standard error of fit or prediction (on the scale of the linear predictors). Default value is FALSE.

• quantile: for type equal to quantile, the quantiles of the distribution. Default is (1:9)/10.

• eval_time: for type equal to "survival" or "hazard", the time points at which the survival probability or hazard is estimated.

A single character string for the type of model. The only possible value for this model is "regression".

The number of features that will be retained in the final model, including the intercept.

The highest possible interaction degree.

The pruning method.

lag periods

external regressors

forecast horizon

selected features

A dataframe of xreg (exogenous regressors)

A numeric vector of values to fit

The number of features that will be retained in the final model, including the intercept.

The highest possible interaction degree.

The pruning method.

lag periods

external regressors

forecast horizon

selected features

model object

input data to predict

Additional parsnip-related options, depending on the value of type. Arguments to the underlying model's prediction function cannot be passed here (use the opts argument instead). Possible arguments are:

• interval: for type equal to "survival" or "quantile", should interval estimates be added, if available? Options are "none" and "confidence".

• level: for type equal to "conf_int", "pred_int", or "survival", this is the parameter for the tail area of the intervals (e.g. confidence level for confidence intervals). Default value is 0.95.

• std_error: for type equal to "conf_int" or "pred_int", add the standard error of fit or prediction (on the scale of the linear predictors). Default value is FALSE.

• quantile: for type equal to quantile, the quantiles of the distribution. Default is (1:9)/10.

• eval_time: for type equal to "survival" or "hazard", the time points at which the survival probability or hazard is estimated.

model object

input data to predict

model object

input data to predict

model object

input data to predict

model object

input data to predict

model object

input data to predict

model object

input data to predict

Run info using set_run_info()

A standard data frame, tibble, or spark data frame using sparklyr of historical time series data. Can also include external regressors for both historical and future data.

List of column headers within input data to be used to separate individual time series.

The column header formatted as a character value within input data you want to forecast.

The date granularity of the input data. Finn accepts the following as a character string: day, week, month, quarter, year.

Number of periods to forecast into the future.

List of column headers within input data to be used as features in multivariate models.

Date value of when your input_data starts. Default of NULL uses earliest date value in input_data.

Date value of when your input_data ends. Default of NULL uses the latest date value in input_data.

Date value to remove individual time series that don't contain non-zero values after that specified date. Default of NULL is to not remove any time series and attempt to forecast all time series.

Month number of start of fiscal year of input data, aids in building out date features. Formatted as a numeric value. Default of 1 assumes fiscal year starts in January.

If TRUE, cleans missing values. Only impute values for missing data within an existing series, and does not add new values onto the beginning or end, but does provide a value of 0 for said values.

If TRUE, outliers are cleaned and inputted with values more in line with historical data.

Apply box-cox transformation to normalize variance in data

Apply differencing to make data stationary

How the forecast is created. The default of 'bottoms_up' trains models for each individual time series. Value of 'grouped_hierarchy' creates a grouped time series to forecast at while 'standard_hierarchy' creates a more traditional hierarchical time series to forecast, both based on the hts package.

Default of NULL runs no parallel processing and forecasts each individual time series one after another. Value of 'local_machine' leverages all cores on current machine Finn is running on. Value of 'spark' runs time series in parallel on a spark cluster in Azure Databricks/Synapse.

Number of cores to run when parallel processing is set up. Used when running parallel computations on local machine or within Azure. Default of NULL uses total amount of cores on machine minus one. Can't be greater than number of cores on machine minus 1.

If TRUE, log transform target variable before training models.

List of values to use in creating fourier series as features. Default of NULL automatically chooses these values based on the date_type.

List of values to use in creating lag features. Default of NULL automatically chooses these values based on date_type.

List of values to use in creating rolling window features. Default of NULL automatically chooses these values based on date_type.

List of recipes to run on multivariate models that can run different recipes. A value of NULL runs all recipes, but only runs the R1 recipe for weekly and daily date types. A value of "all" runs all recipes, regardless of date type. A list like c("R1") or c("R2") would only run models with the R1 or R2 recipe.

Use a multistep horizon approach when training multivariate models with R1 recipe.

run info using the set_run_info() function.

Number of specific back test folds to run when determining the best model. Default of NULL will automatically choose the number of back tests to run based on historical data size, which tries to always use a minimum of 80% of the data when training a model.

Number of periods to move back for each back test scenario. Default of NULL moves back 1 period at a time for year, quarter, and month data. Moves back 4 for week and 7 for day data.

List of models to run. Default of NULL runs all models.

List of models not to run, overrides values in models_to_run. Default of NULL doesn't turn off any model.

If TRUE, prep for ensemble models.

If TRUE, run principle component analysis on any lagged features to speed up model run time. Default of NULL runs PCA on day and week date types across all local multivariate models, and also for global models across all date types.

number of hyperparameter combinations to test out on validation data for model tuning.

Set seed for random number generator. Numeric value.

Name used to group similar runs under a single experiment name.

Name to distinguish one run of Finn from another. The current time in UTC is appended to the run name to ensure a unique run name is created.

Used to store outputs during a run to other storage services in Azure. Could be a storage container object from the 'AzureStor' package to connect to ADLS blob storage or a OneDrive/SharePoint object from the 'Microsoft365R' package to connect to a OneDrive folder or SharePoint site. Default of NULL will save outputs to the local file system.

String showing what file path the outputs should be written to. Default of NULL will write the outputs to a temporary directory within R, which will delete itself after the R session closes.

String value describing the file type for data outputs. Default will write data frame outputs as csv files. The other option of 'parquet' will instead write parquet files.

String value describing the file type for object outputs. Default will write object outputs like trained models as rds files. The other option of 'qs' will instead serialize R objects as qs files by using the 'qs' package.

Add a unique id to end of run_name based on submission time. Set to FALSE to supply your own unique run name, which is helpful in multistage ML pipelines.

A single character string for the type of model. The only possible value for this model is "regression".

A positive number for the cost of predicting a sample within or on the wrong side of the margin.

A positive number for polynomial degree.

A positive number for the polynomial scaling factor.

A positive number for the epsilon in the SVM insensitive loss function

lag periods

external regressors

forecast horizon

selected features

A dataframe of xreg (exogenous regressors)

A numeric vector of values to fit

A positive number for the cost of predicting a sample within or on the wrong side of the margin.

A positive number for polynomial degree.

A positive number for the polynomial scaling factor.

A positive number for the epsilon in the SVM insensitive loss function

lag periods

external regressors

forecast horizon

selected features

model object

input data to predict

Additional parsnip-related options, depending on the value of type. Arguments to the underlying model's prediction function cannot be passed here (use the opts argument instead). Possible arguments are:

• interval: for type equal to "survival" or "quantile", should interval estimates be added, if available? Options are "none" and "confidence".

• level: for type equal to "conf_int", "pred_int", or "survival", this is the parameter for the tail area of the intervals (e.g. confidence level for confidence intervals). Default value is 0.95.

• std_error: for type equal to "conf_int" or "pred_int", add the standard error of fit or prediction (on the scale of the linear predictors). Default value is FALSE.

• quantile: for type equal to quantile, the quantiles of the distribution. Default is (1:9)/10.

• eval_time: for type equal to "survival" or "hazard", the time points at which the survival probability or hazard is estimated.

A single character string for the type of model. The only possible value for this model is "regression".

A positive number for the cost of predicting a sample within or on the wrong side of the margin.

A positive number for radial basis function.

A positive number for the epsilon in the SVM insensitive loss function.

lag periods

external regressors

forecast horizon

selected features

A dataframe of xreg (exogenous regressors)

A numeric vector of values to fit

A positive number for the cost of predicting a sample within or on the wrong side of the margin.

A positive number for radial basis function.

A positive number for the epsilon in the SVM insensitive loss function

lag periods

external regressors

forecast horizon

selected features

model object

input data to predict

Additional parsnip-related options, depending on the value of type. Arguments to the underlying model's prediction function cannot be passed here (use the opts argument instead). Possible arguments are:

• interval: for type equal to "survival" or "quantile", should interval estimates be added, if available? Options are "none" and "confidence".

• level: for type equal to "conf_int", "pred_int", or "survival", this is the parameter for the tail area of the intervals (e.g. confidence level for confidence intervals). Default value is 0.95.

• std_error: for type equal to "conf_int" or "pred_int", add the standard error of fit or prediction (on the scale of the linear predictors). Default value is FALSE.

• quantile: for type equal to quantile, the quantiles of the distribution. Default is (1:9)/10.

• eval_time: for type equal to "survival" or "hazard", the time points at which the survival probability or hazard is estimated.

run info using the set_run_info() function

If TRUE, run multivariate models on the entire data set (across all time series) as a global model. Can be override by models_not_to_run. Default of NULL runs global models for all date types except week and day.

If TRUE, run models by individual time series as local models.

Recipes to use in global models.

Implement feature selection before model training

If TRUE, allow forecasts to dip below zero.

Default of NULL runs no parallel processing and forecasts each individual time series one after another. 'local_machine' leverages all cores on current machine Finn is running on. 'spark' runs time series in parallel on a spark cluster in Azure Databricks or Azure Synapse.

Run components of forecast process inside a specific time series in parallel. Can only be used if parallel_processing is set to NULL or 'spark'.

Number of cores to run when parallel processing is set up. Used when running parallel computations on local machine or within Azure. Default of NULL uses total amount of cores on machine minus one. Can't be greater than number of cores on machine minus 1.

Set seed for random number generator. Numeric value.

model object

parameters

committees

neighbors

max rules

lag periods

external regressors

forecast horizon

selected features

fresh

extra args passed to cubist

model object

parameters

mixture

penalty

lag periods

external regressors

forecast horizon

selected features

fresh

extra args passed to glmnet

model object

parameters

The number of features that will be retained in the final model, including the intercept.

The highest possible interaction degree.

The pruning method.

lag periods

external regressors

forecast horizon

selected features

fresh

extra args passed to mars

model object

parameters

A positive number for the cost of predicting a sample within or on the wrong side of the margin.

A positive number for polynomial degree.

A positive number for the polynomial scaling factor.

A positive number for the epsilon in the SVM insensitive loss function

lag periods

external regressors

forecast horizon

selected features

fresh

extra args passed to svm_poly

model object

parameters

A positive number for the cost of predicting a sample within or on the wrong side of the margin.

A positive number for radial basis function.

A positive number for the epsilon in the SVM insensitive loss function.

lag periods

external regressors

forecast horizon

selected features

fresh

extra args passed to svm_rbf

model object

parameters

mtry

trees

min_n

tree depth

learn rate

loss reduction

number for the number (or proportion) of data that is exposed to the fitting routine.

The number of iterations without improvement before stopping

lag periods

external regressors

forecast horizon

selected features

fresh

extra args passed to xgboost

A single character string for the type of model. The only possible value for this model is "regression".

mtry

trees

min_n

tree depth

learn rate

loss reduction

number for the number (or proportion) of data that is exposed to the fitting routine.

The number of iterations without improvement before stopping

lag periods

external regressors

forecast horizon

selected features

A dataframe of xreg (exogenous regressors)

A numeric vector of values to fit

An integer for the maximum depth of the tree.

An integer for the number of boosting iterations.

A numeric value between zero and one to control the learning rate.

Subsampling proportion of columns.

Subsampling proportion of columns for each node within each tree. See the counts argument below. The default uses all columns.

A numeric value for the minimum sum of instance weights needed in a child to continue to split.

A number for the minimum loss reduction required to make a further partition on a leaf node of the tree

Subsampling proportion of rows.

A positive number. If on ⁠[0, 1)⁠ the value, validation is a random proportion of data in x and y that are used for performance assessment and potential early stopping. If 1 or greater, it is the number of training set samples use for these purposes.

An integer or NULL. If not NULL, it is the number of training iterations without improvement before stopping. If validation is used, performance is base on the validation set; otherwise the training set is used.

lag periods

external regressors

forecast horizon

selected features

Additional arguments passed to xgboost::xgb.train

model object

input data to predict

Additional arguments passed to predict.xgb.Booster()