tune_nbeats: Function to tune N-BEATS
In vidarsumo/sumots: Helper functions for Sumo Analytics

Function to tune N-BEATS

tune_nbeats(
  id,
  freq,
  recipe,
  splits,
  horizon,
  length,
  cv_slice_limit,
  assess = "12 weeks",
  skip = "4 weeks",
  initial = "12 months",
  epochs = NULL,
  lookback = NULL,
  bagging_size = NULL,
  learn_rate = NULL,
  loss_function = NULL,
  scale = NULL,
  batch_size = NULL
)

`id`	A quoted column name that tracks the GluonTS FieldName "item_id"
`freq`	A pandas timeseries frequency such as "5min" for 5-minutes or "D" for daily.
`recipe`	A gluonts recipe
`horizon`	The forecast horizon
`length`	The number of distinct hyperparameter for each tunable parameter except loss_function which is set to MASE
`cv_slice_limit`	How many slice/folds in the tsCV
`assess`	The number of samples used for each assessment resample
`skip`	A integer indicating how many (if any) additional resamples to skip to thin the total amount of data points in the analysis resample.
`initial`	The number of samples used for analysis/modeling in the initial resample.
`epochs`	Number of epochs
`lookback`	Lookback length. If NULL, will be randomly chosen
`bagging_size`	The number of models that share the parameter combination of 'context_length' and 'loss_function'.
`learn_rate`	Learning rate
`loss_function`	Any of MASE, MAPE sMAPE. Defaults to MASE when loss_function = NULL
`scale`	Scales numeric data by id group using mean = 0, standard deviation = 1 transformation.
`batch_size`	Number of examples in each batch