plot.md

ELAPS Report plotting

The matplotlib-based module elaps.plot provides plot() that plots a series of metric data sets as produced by Report.evaluate(). If any of the provided data sets contains a range, it produces a line-plot, otherwise a bar-plot.

Table of Contents generated with DocToc

• Input
  • datas
  • stat_names
  • colors
  • styles
  • xlabel
  • ylabel
  • legendargs
  • figure
• Output

Input

plot() takes the following arguments, all of which but the first are optional.

datas

(required)
The datasets to be plotted as a list of two-element tuples consisting of the legend entry and the data set as returned by Report.apply_metric().

stat_names

(default: ["med"])
A list of statistics to be plotted. May contain "min", "med", "max", "avg", "std", and "all".

"min" through "avg" on their own are plotted as lines (or simple bars); when both "min" and "max" are present, the range between them ("min-max") is filled; "std" fills the range between of the average +/- one standard deviation; "all" plots all data points as markers.

colors

(default: built-in color set)
A list of colors for the datasets in the same order.

styles

(default: built-in styles)
A dict of style options for different statistics that overwrite the built-in options. Relevant: the statistics names, "min-max" and "legend". Allowed values: keyword arguments (dict) for matplotlib's plot() method (fill_between() for "min-max" and "std").

xlabel

(default: no label)
The plots x-axis label. Ignored for bar-plots.

ylabel

(default: no label)
The plots y-axis label.

legendargs

(default: {}) Keyword arguments (dict) for matplotlib's legend() method (e.g., for custom legend placement).

figure

(default: pyplot.gcf())
The matplotlib Figure in which to plot.

Output

plot() returns a matplotlib Figure that can be further modified or exported via its savefig() method.