LECA.analyze.predict_conductivity_from_arrhenius_objectives

LECA.analyze.predict_conductivity_from_arrhenius_objectives(x_in: DataFrame, wf: WorkFlow, model: str | List[str], beta_0: float, log: bool = False, min_max: bool = False) → DataFrame

Predict the ionic conductivity for given electrolyte compositions at a given temperature.

Parameters:

  • x_in (pd.DataFrame) – DataFrame of input feature vectors, matching the DataFrame format for WorkFlow.

  • wf (WorkFlow) – LECA WorkFlow object containing trained models for Arrhenius objective functions S0, S1 and S2.

  • model (Union[str, List[str]]) – String or list of [string, string, string]. If a single string is provided, the 3 models with that name are drawn from the WorkFlow for predicting S0/S1/S2 respectively. Otherwise, the list position corresponds to the model name selected from the WorkFlow. I.e, the model name passed in model[0] will be used to predict S0, model[1] -> S1 and model[2] -> S2.

  • beta_0 (float) – beta_0 value used for the Arrhenius surrogate model. This value should have been stored during the transformation of measurement data into the Arrhenius surrogate model objectives (See: prep.arrhenius(), prep.direct_sample_arrhenius()).

  • log (bool) –

    If log=True return log_{10}(conductivity). If log=False return conductivity.

    Default value False

  • min_max (bool) –

    Whether to use the min_max method to estimate uncertainty, or MAPIE with conformity scores. min_max:True returns the standard deviations of the predictions from all the bootstrapped models for each point. min_max:False uses the MAPIE uncertainty estimation outlined in: Mapie jackknife+-AB This parameter is moot for GPR models.

    Default value False.

Returns:

DataFrame with the following columns structure of input features and conductivity predictions (and their one-sigma uncertainty). x_i here signifies each feature dimension in x_in.

inverse temperature

x_1

x_2

x…

conductivity

conducivity_std

Return type:

pd.DataFrame