fitbenchmarking.results_processing.local_min_table module

compare table

class fitbenchmarking.results_processing.local_min_table.LocalMinTable(results, best_results, options, group_dir, pp_locations, table_name)

Bases: fitbenchmarking.results_processing.base_table.Table

The local min results shows a True or False value together with \(\frac{|| J^T r||}{||r||}\). The True or False indicates whether the software finds a minimum with respect to the following criteria:

  • \(||r|| \leq\) RES_TOL,

  • \(|| J^T r|| \leq\) GRAD_TOL,

  • \(\frac{|| J^T r||}{||r||} \leq\) GRAD_TOL,

where \(J\) and \(r\) are the Jacobian and residual of \(f(x, p)\), respectively. The tolerances can be found in the results object.

display_str(value)

Combine the boolean value from variable local_min with the normalised residual

Parameters

value (bool, float) – Whether the minimizer found a local minimizer and the \(\frac{|| J^T r||}{||r||}\) value

Returns

string representation of the value for display in the table.

Return type

str

get_value(result)

Gets the main value to be reported in the tables for a given result

Note that the first value (relative chi_sq) will be used in the default colour handling.

Parameters

result (FittingResult) – The result to generate the values for.

Returns

Whether the minimizer found a local minimizer (under the tests specified above) and \(\frac{|| J^T r||}{||r||}\)

Return type

bool, float

save_colourbar(fig_dir, n_divs=2, sz_in=None) → str

Override default save_colourbar as there are only 2 possible divisions of the colour map (true or false).

Parameters

  • fig_dir (str) – path to figures directory

  • n_divs (int, Fixed to 2) – Unused number of divisions of shading in colourbar

  • sz_in (list[float] - 2 elements) – dimensions of png in inches [width, height]

Returns

The relative path to the colourbar image.

Return type

str

static vals_to_colour(vals, cmap, cmap_range, log_ulim)

Converts an array of values to a list of hexadecimal colour strings using sampling from a matplotlib colourmap according to whether a minimum was found.

Set to the bottom of the range if minimum was found, otherwise set to the top of the range.

Parameters

  • vals (list[float]) – values in the range [0, 1] to convert to colour strings

  • cmap (matplotlib colourmap object) – matplotlib colourmap

  • cmap_range (list[float], 2 elements) – values in range [0, 1] for colourmap cropping

  • log_ulim (float) – Unused log10 of worst shading cutoff value

Returns

colours as hex strings for each input value

Return type

list[str]