Core Module Documentation

class neuralprophet.configure.AR(n_lags: int, ar_reg: Optional[float] = None)
regularize(weights, original=False)

Regularization of AR coefficients

Parameters
  • weights (torch.Tensor) – Model weights to be regularized towards zero

  • original (bool) – Do not penalize non-zeros

Returns

Regularization loss

Return type

numeric

class neuralprophet.configure.AllSeason(mode: str = 'additive', computation: str = 'fourier', reg_lambda: float = 0, yearly_arg: (<class 'str'>, <class 'bool'>, <class 'int'>) = 'auto', weekly_arg: (<class 'str'>, <class 'bool'>, <class 'int'>) = 'auto', daily_arg: (<class 'str'>, <class 'bool'>, <class 'int'>) = 'auto')
class neuralprophet.configure.Covar(reg_lambda: float, as_scalar: bool, normalize: (<class 'bool'>, <class 'str'>))
class neuralprophet.configure.Event(lower_window: int, upper_window: int, reg_lambda: float, mode: str)
class neuralprophet.configure.Holidays(country: str, lower_window: int, upper_window: int, mode: str = 'additive', reg_lambda: float = None)
class neuralprophet.configure.Model(num_hidden_layers: int, d_hidden: int)
class neuralprophet.configure.Normalization(normalize: str, global_normalization: bool, global_time_normalization: bool, unknown_data_normalization: bool, local_data_params: dict = None, global_data_params: dict = None)
class neuralprophet.configure.Regressor(reg_lambda: float, normalize: str, mode: str)
class neuralprophet.configure.Season(resolution: int, period: float, arg: str)
class neuralprophet.configure.Train(learning_rate: (<class 'float'>, None), epochs: (<class 'int'>, None), batch_size: (<class 'int'>, None), loss_func: (<class 'str'>, <class 'torch.nn.modules.loss._Loss'>, 'typing.Callable'), optimizer: (<class 'str'>, <class 'torch.optim.optimizer.Optimizer'>), newer_samples_weight: float = 1.0, newer_samples_start: float = 0.0, reg_delay_pct: float = 0.5, reg_lambda_trend: float = None, trend_reg_threshold: (<class 'bool'>, <class 'float'>) = None, reg_lambda_season: float = None)
class neuralprophet.configure.Trend(growth: str, changepoints: list, n_changepoints: int, changepoints_range: float, trend_reg: float, trend_reg_threshold: (<class 'bool'>, <class 'float'>))