coolest.api.plotting

Classes

ModelPlotter	Create pyplot panels from a lens model stored in the COOLEST format.
MultiModelPlotter	Wrapper around a set of ModelPlotter instances to produce panels that
ParametersPlotter	Handles plot of analytical models in a comparative way

Module Contents

class ModelPlotter(coolest_object, coolest_directory=None, color_bad_values='#222222')

Create pyplot panels from a lens model stored in the COOLEST format.

Parameters:

• coolest_object (COOLEST) – COOLEST instance

• coolest_directory (str, optional) – Directory which contains the COOLEST template, by default None

• color_bad_values (str, optional) – Color assigned to NaN values (typically negative values in log-scale), by default ‘#111111’ (dark gray)

coolest

cmap_flux

cmap_mag = 'viridis'

cmap_conv = 'cividis'

cmap_res = 'RdBu_r'

plot_data_image(ax, title=None, norm=None, cmap=None, xylim=None, neg_values_as_bad=False, add_colorbar=True, add_scalebar=True, scalebar_size=1)

plt.imshow panel with the data image

plot_surface_brightness(ax, title=None, coordinates=None, extent_irreg=None, norm=None, cmap=None, xylim=None, neg_values_as_bad=True, plot_points_irreg=False, add_colorbar=True, add_scalebar=False, scalebar_size=0.4, kwargs_light=None, plot_caustics=None, caustics_color='white', caustics_alpha=0.5, coordinates_lens=None, kwargs_lens_mass=None)

plt.imshow panel showing the surface brightness of the (unlensed) lensing entity selected via kwargs_light (see ComposableLightModel docstring)

plot_model_image(ax, title=None, norm=None, cmap=None, xylim=None, neg_values_as_bad=False, add_colorbar=True, add_scalebar=True, scalebar_size=1, auto_selection=False, kwargs_lens_mass=None, kwargs_lens_light=None, kwargs_source=None, **model_image_kwargs)

plt.imshow panel showing the surface brightness of the (lensed) selected lensing entities (see ComposableLensModel docstring)

plot_model_residuals(ax, title=None, mask=None, norm=None, cmap=None, xylim=None, add_chi2_label=False, chi2_fontsize=12, kwargs_source=None, kwargs_lens_mass=None, kwargs_lens_light=None, add_colorbar=True, add_scalebar=True, scalebar_size=1, **model_image_kwargs)

plt.imshow panel showing the normalized model residuals image

plot_convergence(ax, title=None, coordinates=None, norm=None, cmap=None, xylim=None, neg_values_as_bad=False, add_colorbar=True, add_scalebar=True, scalebar_size=1, kwargs_lens_mass=None)

plt.imshow panel showing the 2D convergence map associated to the selected lensing entities (see ComposableMassModel docstring)

plot_convergence_diff(ax, reference_map, title=None, relative_error=True, norm=None, cmap=None, xylim=None, coordinates=None, add_colorbar=True, add_scalebar=True, scalebar_size=1, kwargs_lens_mass=None, plot_crit_lines=False, crit_lines_color='black', crit_lines_alpha=0.5)

plt.imshow panel showing the 2D convergence map associated to the selected lensing entities (see ComposableMassModel docstring)

plot_magnification(ax, title=None, norm=None, cmap=None, xylim=None, add_colorbar=True, add_scalebar=True, scalebar_size=1, coordinates=None, kwargs_lens_mass=None)

plt.imshow panel showing the 2D magnification map associated to the selected lensing entities (see ComposableMassModel docstring)

plot_magnification_diff(ax, reference_map, title=None, relative_error=True, norm=None, cmap=None, xylim=None, add_colorbar=True, add_scalebar=True, scalebar_size=1, coordinates=None, kwargs_lens_mass=None)

plt.imshow panel showing the (absolute or relative) difference between 2D magnification maps

class MultiModelPlotter(coolest_objects, coolest_directories=None, **kwargs_plotter)

Wrapper around a set of ModelPlotter instances to produce panels that consistently compare different models, evaluated on the same coordinates systems.

Parameters:

• coolest_objects (list) – List of COOLEST instances

• coolest_directories (list, optional) – List of directories corresponding to each COOLEST instance, by default None

• kwargs_plotter (dict, optional) – Additional keyword arguments passed to ModelPlotter

num_models

plotter_list = []

plot_surface_brightness(axes, **kwargs)

plot_data_image(axes, **kwargs)

plot_model_image(axes, **kwargs)

plot_model_residuals(axes, **kwargs)

plot_convergence(axes, **kwargs)

plot_magnification(axes, **kwargs)

plot_convergence_diff(axes, *args, **kwargs)

plot_magnification_diff(axes, *args, **kwargs)

class ParametersPlotter(parameter_id_list, coolest_objects, coolest_directories=None, coolest_names=None, ref_coolest_objects=None, ref_coolest_directories=None, ref_coolest_names=None, posterior_bool_list=None, colors=None, linestyles=None, add_multivariate_margin_samples=False, num_samples_per_model_margin=5000)

Handles plot of analytical models in a comparative way

Parameters:

• parameter_id_list (array) – A list of parameter unique ids obtained from lensing entities. Their order determines the order of the plot panels.

• coolest_objects (array) – A list of coolest objects that have a chain file associated to them.

• coolest_directories (array) – A list of paths matching the coolest files in ‘chain_objs’.

• coolest_names (array, optional) – A list of labels for the coolest models in the ‘chain_objs’ list. Must have the same order as ‘chain_objs’.

• ref_coolest_objects (array, optional) – A list of coolest objects that will be used as point estimates.

• ref_coolest_directories (array) – A list of paths matching the coolest files in ‘point_estimate_objs’.

• ref_coolest_names (array, optional) – A list of labels for the models in the ‘point_estimate_objs’ list. Must have the same order as ‘point_estimate_objs’.

• posterior_bool_list (list, optional) – List of bool to toggle errorbars on point-estimate values

• colors (list, optional) – List of pyplot color names to associate to each coolest model.

• linestyles (list, optional) – List of pyplot linesyles to associate to each coolest model.

• add_multivariate_margin_samples (bool, optional) – If True, will append to the list of compared models a new chain that is resampled from the multi-variate normal distribution, where its covariance matrix is computed from the marginalization of all samples from all models. By default False.

• num_samples_per_model_margin (int, optional) – Number of samples to (randomly) draw from each model samples to concatenate before estimating the multi-variate normal marginalization.

parameter_id_list

coolest_objects

coolest_directories = None

coolest_names = None

ref_coolest_objects = None

ref_coolest_directories = None

ref_coolest_names = None

ref_file_names = None

num_models

num_params

colors = None

linestyles = None

ref_linestyles = ['--', ':', '-.', '-']

ref_markers = ['s', '^', 'o', '*']

init_getdist(shift_sample_list=None, settings_mcsamples=None, add_multivariate_margin_samples=False)

Initializes the getdist plotter.

Parameters:

• shift_sample_list (dict) – Dictionary keyed by parameter ID to apply a uniform additive shift to all samples of that parameters posterior distribution.

• settings_mcsamples (dict, optional) – Keyword arguments passed as the settings argument of getdist.MCSamples, by default None

Raises:

ValueError – If the csv file containing samples is is not coma (,) separated.

get_mcsamples_getdist(with_margin=False)

get_margin_mcsamples_getdist()

plot_triangle_getdist(filled_contours=True, angles_range=None, linewidth_hist=2, linewidth_cont=2, linewidth_margin=4, marker_linewidth=2, marker_size=15, axes_labelsize=None, legend_fontsize=None, **subplot_kwargs)

Corner array of subplots using getdist.triangle_plot method.

Parameters:

• subplot_size (int, optional) – Size of the getdist plot, by default 1

• filled_contours (bool, optional) – Wether or not to fill the 2D contours, by default True

• angles_range (_type_, optional) – Restrict the range of angle (containing ‘phi’ in their name) parameters, by default None

• linewidth_hist (int, optional) – Line width for 1D histograms, by default 2

• linewidth_cont (int, optional) – Line width for 2D contours, by default 1

• marker_size (int, optional) – Size of the reference (scatter) markers on 2D contours plots, by default 15

Returns:

Instance of GetDistPlotter corresponding to the figure

Return type:

GetDistPlotter

plot_rectangle_getdist(x_param_ids, y_param_ids, subplot_size=1, legend_ncol=None, legend_fontsize=None, filled_contours=True, linewidth=1, marker_size=15, axes_labelsize=None, **subplot_kwargs)

Array of (2D contours) subplots using getdist.rectangle_plot method.

Parameters:

• subplot_size (int, optional) – Size of the getdist plot, by default 1

• filled_contours (bool, optional) – Wether or not to fill the 2D contours, by default True

• linewidth (int, optional) – Line width for 2D contours, by default 1

• marker_size (int, optional) – Size of the reference (scatter) markers on 2D contours plots, by default 15

• legend_ncol (number of columns in the legend)

Returns:

Instance of GetDistPlotter corresponding to the figure

Return type:

GetDistPlotter

plot_1d_getdist(num_columns=None, legend_ncol=None, legend_fontsize=None, axes_labelsize=None, linewidth=1, **subplot_kwargs)

Array of 1D histogram subplots using getdist.plots_1d method.

Parameters:

• subplot_size (int, optional) – Size of the getdist plot, by default 1

• linewidth (int, optional) – Line width for 2D contours, by default 1

• marker_size (int, optional) – Size of the reference (scatter) markers on 2D contours plots, by default 15

• legend_ncol (int, optional) – number of columns in the legend

• num_columns (int, optional) – number of columns of the subplot array

Returns:

Instance of GetDistPlotter corresponding to the figure

Return type:

GetDistPlotter

plot_source(idx_file=0)

plot_lens(idx_file=0)

plotting_routine(param_dict, idx_file=0)

plot the parameters

INPUT

param_dict: dict, organized dictonnary with all parameters results of the different files idx_file: int, chooses the file on which the choice of plotted parameters will be made (not very clear: basically in file 0 you may have a sersic fit and in file 1 sersic+shapelets. If you choose

idx_file=0, you will plot the sersic results of both file. If you choose idx_file=1, you will plot all the sersic and shapelets parameters when available)