src.driver.visualization

Visualization and plotting logic for MMM V2 workflow.

This module handles all aspects of result visualization including model diagnostics, posterior plots, and component contributions.

Module Contents

class src.driver.visualization.ResultVisualizer(driver)

Handles visualization of MMM results for V2 architecture.

This class manages all plotting and visualization operations for model results, diagnostics, and component analysis.

generate_all() → None

Visualizes all results of the model.

plot_model_structure() → graphviz.Digraph

Plots the model structure.

Returns:

Graphviz diagram of model structure.

Raises:

ValueError – If model not available or PyMC model not found.

plot_model_trace() → matplotlib.figure.Figure

Plots the trace of the model parameters.

Returns:

Matplotlib figure with trace plots.

Raises:

ValueError – If model not fitted or idata not available.

plot_components_contributions() → matplotlib.figure.Figure

Plots the contribution of each model component over time.

Returns:

Matplotlib figure with component contributions.

Raises:

• ValueError – If model not available.

• AttributeError – If model doesn’t have plot_components_contributions method.

plot_posterior_predictions() → matplotlib.figure.Figure

Plots the in-sample and out-of-sample posterior predictions.

Returns:

Matplotlib figure with predictions vs actuals.

Raises:

ValueError – If model or results directory not available.

plot_waterfall_components_decomposition(model_to_plot: src.core.base.BaseMMM | None = None, original_scale: bool = True, figsize: Tuple[int, int] = (14, 7), **kwargs: Any) → matplotlib.figure.Figure

Plots the waterfall decomposition of the model components.

Parameters:

• model_to_plot – Model to plot (defaults to driver’s model).

• original_scale – Whether to plot in original scale.

• figsize – Figure size tuple.

• **kwargs – Additional arguments for plotting.

Returns:

Matplotlib figure with waterfall decomposition.

Raises:

ValueError – If no model available.

plot_posterior_distributions(results_dir_override: str | None = None, filename_prefix: str = 'media_spend_posterior') → pandas.DataFrame | None

Plots posterior distributions for all media channels.

Parameters:

• results_dir_override – Specific directory to save plots.

• filename_prefix – Prefix for the saved plot filename.

Returns:

DataFrame with statistical summaries of posteriors, or None if plotting fails.

display_image(image_filename: str) → Any | None

Displays an image file located in the results directory.

Parameters:

image_filename – Filename of the image to display.

Returns:

IPython Image object if available, None otherwise.

describe_all_media_spend() → Any

Describes overall media spend.

Returns:

Description of media spend.

plot_all_media_spend() → Any

Plots overall media spend.

Returns:

Plot of media spend.