nixtla_client_cross_validation {nixtlar}

Perform cross validation with 'TimeGPT'.

Description

Perform cross validation with 'TimeGPT'.

Usage

nixtla_client_cross_validation(
  df,
  h = 8,
  freq = NULL,
  id_col = NULL,
  time_col = "ds",
  target_col = "y",
  X_df = NULL,
  level = NULL,
  n_windows = 1,
  step_size = NULL,
  finetune_steps = 0,
  finetune_loss = "default",
  clean_ex_first = TRUE,
  model = "timegpt-1"
)

Arguments

df

A tsibble or a data frame with time series data.

h

Forecast horizon.

freq

Frequency of the data.

id_col

Column that identifies each series.

time_col

Column that identifies each timestep.

target_col

Column that contains the target variable.

X_df

A tsibble or a data frame with future exogenous variables.

level

The confidence levels (0-100) for the prediction intervals.

n_windows

Number of windows to evaluate.

step_size

Step size between each cross validation window. If NULL, it will equal the forecast horizon (h).

finetune_steps

Number of steps used to finetune 'TimeGPT' in the new data.

finetune_loss

Loss function to use for finetuning. Options are: "default", "mae", "mse", "rmse", "mape", and "smape".

clean_ex_first

Clean exogenous signal before making the forecasts using 'TimeGPT'.

model

Model to use, either "timegpt-1" or "timegpt-1-long-horizon". Use "timegpt-1-long-horizon" if you want to forecast more than one seasonal period given the frequency of the data.

Value

A tsibble or a data frame with 'TimeGPT”s cross validation result.

Examples

## Not run: 
  nixtlar::nixtla_set_api_key("YOUR_API_KEY")
  df <- nixtlar::electricity
  fcst <- nixtlar::nixtla_client_cross_validation(df, h = 8, id_col = "unique_id", n_windows = 5)

## End(Not run)