xopto.dataset.validation.mcml_comparison module¶

compare(test_dir: str, reference_dir: str, out_dir: Optional[str] = None, verbose: bool = False)[source]¶

Compares the results of two “MCML comparison” datasets that include data on several configurations computed with layered (mcml) and voxelized (mcvox) Monte Carlo simulator cores.

Parameters

test_dir (str) – Path to the root directory of the test dataset.
reference_dir (str) – Path to the root directory of the reference dataset.
out_dir (str) – Optional directory for producing error plots. The error plots follow the directory structure of the dataset.
verbose (bool) – Enable verbose reporting to stdout.

Returns

results – A collection of Relative Root-Mean-Square error (RRMSE) and Mean Relative Error (MRE) values for all the dataset configurations defined in the CONFIG dict. The MRE and RRMSE values are given in multidimensional numpy arrays. The following code illustrates the organization of the returned data. note that the fields are fully expanded only for configuration 1-layer-100mm simulated with the Monte Carlo core for layered media (mcml).

{
    'mcml': {
        '1-layer-100mm': {
            'transmittance': {'mre': values, 'rrmse': values},
            'reflectance': {'mre': values, 'rrmse': values},
            'total_transmittance': {'mre': values, 'rrmse': values},
            'total_reflectance': {'mre': values, 'rrmse': values},
            'fluence_data': {'mre': values, 'rrmse': values},
            'indexing': ('mua', 'musr', 'g')
            'mua': mua_values,
            'musr': msr_values,
            'g': g_values
        },
        '1-layer-1mm': {...},
        '2-layer-100um-1mm': {...}
},
    'mcvox': {
        '1-layer-100mm': {...},
        '1-layer-1mm': {...},
        '2-layer-100um-1mm': {...}
    }
}

Return type

dict