plaid.post.bisect¶

Utiliy function to plot bisect graphs comparing predictions vs. targets dataset.

Functions¶

`prepare_datasets`(→ tuple[dict, dict, list[str]])	Prepare datasets for comparison.
`is_dvipng_available`(→ bool)	Check if dvipng is available on the system for matplotlib figures.
`plot_bisect`(→ None)	Plot a bisect graph comparing predictions vs. targets dataset.

Module Contents¶

prepare_datasets(ref_dataset: plaid.Dataset, pred_dataset: plaid.Dataset, problem_definition: plaid.ProblemDefinition, verbose: bool = False) → tuple[dict, dict, list[str]][source]¶

Prepare datasets for comparison.

Parameters:

ref_dataset (Dataset) – The reference dataset.
pred_dataset (Dataset) – The predicted dataset.
problem_definition (ProblemDefinition) – The common problem for the reference and predicted dataset
verbose (bool, optional) – Verbose mode. Defaults to False.

Returns:

A tuple containing dictionaries of reference and predicted scalar values, and a list of scalar names.

Return type:

tuple[dict[str, list[float]], dict[str, list[float]], list[str]]

is_dvipng_available(verbose: bool) → bool[source]¶

Check if dvipng is available on the system for matplotlib figures.

Returns:: True if dvipng is available, False otherwise.
Return type:: bool

Plot a bisect graph comparing predictions vs. targets dataset.

Parameters:

ref_dataset (Dataset | str | Path) – The reference dataset or its file path.
pred_dataset (Dataset | str | Path) – The predicted dataset or its file path.
problem_def (ProblemDefinition | str | Path) – The common problem for the reference and predicted dataset
scalar (str | int) – The name of the scalar to study or its index.
save_file_name (str, optional) – Figure name when saving to PNG format. Defaults to “bissec_plots”.
verbose (bool, optional) – Verbose mode. Defaults to False.

Raises:

KeyError – If the provided scalar name is not part of the dataset.