plot_pipeline_parameters ======================== .. py:module:: plot_pipeline_parameters .. autoapi-nested-parse:: Pipeline parameter analysis plotting module for ML results analysis. Focuses on visualizing the impact of data transformations and pipeline settings on model performance. Classes ------- .. autoapisummary:: plot_pipeline_parameters.PipelineParameterPlotter Module Contents --------------- .. py:class:: PipelineParameterPlotter(data: pandas.DataFrame) Initializes the PipelineParameterPlotter. :param data: Results DataFrame, must contain columns for pipeline parameters and performance metrics. :type data: pd.DataFrame :raises ValueError: If no pipeline parameter columns are found in the data. .. py:attribute:: data .. py:attribute:: clean_data .. py:attribute:: categorical_params :value: ['resample', 'scale', 'param_space_size', 'percent_missing'] .. py:attribute:: continuous_params .. py:attribute:: available_categorical .. py:attribute:: available_continuous .. py:method:: plot_categorical_parameters(metric: str = 'auc', figsize: Optional[Tuple[int, int]] = None) -> None Creates box plots for categorical pipeline parameters. :param metric: The performance metric to plot. Defaults to 'auc'. :type metric: str, optional :param figsize: Figure size for the plot. Defaults to None. :type figsize: Optional[Tuple[int, int]], optional :raises ValueError: If the specified metric is not found in the data. .. py:method:: plot_continuous_parameters(metric: str = 'auc', figsize: Optional[Tuple[int, int]] = None) -> None Creates scatter plots for continuous pipeline parameters. :param metric: The performance metric to plot. Defaults to 'auc'. :type metric: str, optional :param figsize: Figure size for the plot. Defaults to None. :type figsize: Optional[Tuple[int, int]], optional :raises ValueError: If the specified metric is not found in the data.