SALib.plotting package#

Submodules#

SALib.plotting.bar module#

SALib.plotting.bar.plot(Si_df, ax=None)[source]#

Create bar chart of results.

Examples

>>> from SALib.plotting.bar import plot as barplot
>>> from SALib.test_functions import Ishigami
>>>
>>> # See README for example problem specification
>>>
>>> X = saltelli.sample(problem, 512)
>>> Y = Ishigami.evaluate(X)
>>> Si = sobol.analyze(problem, Y, print_to_console=False)
>>> total, first, second = Si.to_df()
>>> barplot(total)

Parameters:: Si_df (*)
Returns:: * ax
Return type:: matplotlib axes object

SALib.plotting.hdmr module#

Created on Dec 20, 2019

@author: @sahin-abdullah

This submodule produces two different figures: (1) emulator vs simulator, (2) regression lines of first order component functions

SALib.plotting.hdmr.plot(Si)[source]#

SALib.plotting.heatmap module#

SALib.plotting.heatmap.heatmap(sp: Dict, metric: str, index: str, title: str = None, ax=None)[source]#

Plot a heatmap of the target metric.

Parameters:

sp (object, SALib ProblemSpec)
metric (str, metric to plot. Defaults to first metric/result output if None.)
index (str, sensitivity indices to plot (‘S1’, ‘ST’, etc). Displays all if None.)
title (str, plot title to use)
ax (axes object, matplotlib axes object to assign figure to.)

Returns:

Return type:

matplotlib axes object

SALib.plotting.morris module#

Created on 29 Jun 2015

@author: @willu47

This module provides the basic infrastructure for plotting charts for the Method of Morris results

The procedures should build upon and return an axes instance:

import matplotlib.pyplot as plt
Si = morris.analyze(problem, param_values, Y, conf_level=0.95,
                    print_to_console=False, num_levels=10)

# set plot style etc.
fig, ax = plt.subplots(1)
p = SALib.plotting.morris.horizontal_bar_plot(ax, Si, {'marker':'x'})
p.show()

SALib.plotting.morris.covariance_plot(ax, Si, opts=None, unit='')[source]#: Plots mu* against sigma or the 95% confidence interval

SALib.plotting.morris.horizontal_bar_plot(ax, Si, opts=None, sortby='mu_star', unit='')[source]#: Updates a matplotlib axes instance with a horizontal bar plot of mu_star, with error bars representing mu_star_conf.

SALib.plotting.morris.sample_histograms(fig, input_sample, problem, opts=None)[source]#: Plots a set of subplots of histograms of the input sample

SALib.plotting.shapley module#

Bar chart for Shapley effects.

Raw Shapley effects (output-variance units) and their normalized shares (unit interval, summing to one) sit on incompatible scales, so plotting all four shapley/shapley_conf/shapley_normalized/ shapley_normalized_conf columns together in a single bar chart (the default behavior inherited from ResultDict.plot) makes the smaller series unreadable. This module plots one pair at a time instead, defaulting to the normalized shares since those are what’s usually of interest.

SALib.plotting.shapley.plot(Si, ax=None, normalized=True)[source]#

Plot Shapley effects as a bar chart.