prepare_plot_data

bayesflow.utils.prepare_plot_data(estimates: Mapping[str, ndarray] | ndarray, targets: Mapping[str, ndarray] | ndarray, variable_keys: Sequence[str] = None, variable_names: Sequence[str] = None, num_col: int = None, num_row: int = None, figsize: tuple = None, stacked: bool = False, default_name: str = 'v') → Mapping[str, Any]

Procedural wrapper that encompasses all preprocessing steps, including shape-checking, parameter name generation, layout configuration, figure initialization, and collapsing of axes.

Parameters:

estimatesdict[str, ndarray] or ndarray

The model-generated predictions or estimates, which can take the following forms:

• ndarray of shape (num_datasets, num_variables)

  Point estimates for each dataset, where num_datasets is the number of datasets and num_variables is the number of variables per dataset.

• ndarray of shape (num_datasets, num_draws, num_variables)

  Posterior samples for each dataset, where num_datasets is the number of datasets, num_draws is the number of posterior draws, and num_variables is the number of variables.

targetsdict[str, ndarray] or ndarray, optional (default = None)

Ground truth values corresponding to the estimates. Must match the structure and dimensionality of estimates in terms of first and last axis.

variable_keyslist or None, optional, default: None

Select keys from the dictionary provided in samples. By default, select all keys.

variable_namesSequence[str], optional (default = None)

Optional variable names to act as a filter if dicts provided or actual variable names in case of array args

num_colint

Number of columns for the visualization layout

num_rowint

Number of rows for the visualization layout

figsizetuple, optional, default: None

Size of the figure adjusting to the display resolution

stackedbool, optional, default: False

Whether the plots are stacked horizontally

default_namestr, optional (default = “v”)

The default name to use for estimates if None provided