CondAEModel.forward_evaluation

CondAEModel.forward_evaluation(data: ndarray | torch.Tensor, format_out: str = 'df', input_transformed: bool = False, return_untransformed: bool = False) → DataFrame | ndarray | List[List] | Dict | List[Dict] | torch.Tensor[source]

Wrapper function of CondAEModel.predict_y() to evaluate the model on the provided data, and return the output in the desired format.

Parameters:

  • data (Union[pd.DataFrame, np.ndarray, List[List], Dict, List[Dict], torch.Tensor]) – Input data to evaluate in the surrogate model

  • format_out (str, optional, default=”df”) – The format for the returned output. The possible formats are [“dict”, “dict_list”, “df_per_obj”, “df”, “array”, “torch”, “list”], and default is “df”.

  • input_transformed (bool, optional, default=False) – If True, the input data is assumed to be already transformed, and no transformation is applied before evaluating the model

  • return_untransformed (bool, optional, default=False) – If True, the output is returned in the original space, by applying the inverse transformation.

Returns:

Union[pd.DataFrame, np.ndarray, List[List], Dict, List[Dict], torch.Tensor] – The predictions in the desired format, based on the format_out argument, transformed or in original space based on the return_untransformed argument.