mcPlot

Standard plotting class of Matthias Cuntz.

It has the same functionality as the old mc_template.py by Matthias Cuntz but uses the object-oriented approach of st_template.py of Stephan Thober.

It allows plotting on screen, into PDF and PNG files, as well as in HTML files as a wrapper for PNG images or using hvplot.

It is optimised for publication ready plots, either on white or black background.

The simplest way to use mcPlot is to extend the class:

import numpy as np
from pyjams import mcPlot

class PlotIt(mcPlot):
    def __init__(self, *args, **kwargs):
        super().__init__(*args, **kwargs)
        # change e.g. colors
        self.lcol1 = 'cyan'
        # reset global values after colour changes, etc.
        self.set_matplotlib_rcparams()

    def read_data(self):
        # do something
        self.dat = np.arange(10)

    def plot_fig_1(self):
        import matplotlib.pyplot as plt

        self.ifig += 1
        fig = plt.figure(self.ifig)

        sub = fig.add_axes([0.125, 0.667, 0.3375, 0.233])

        larr = sub.plot(self.dat)
        plt.setp(larr[-1], linestyle='-', linewidth=self.lwidth,
                 marker='', color=self.lcol1)

        self.plot_save(fig)

    def plot_fig_2(self):
        import matplotlib.pyplot as plt

        self.ifig += 1
        fig = plt.figure(self.ifig)

        sub = fig.add_axes([0.125, 0.667, 0.3375, 0.233])

        larr = sub.plot(2*self.dat)
        plt.setp(larr[-1], linestyle='-', linewidth=self.lwidth,
                 marker='', color=self.lcols[-1])

        self.plot_save(fig)

if __name__ == '__main__':
    iplot = PlotIt(desc='Test mcPlot',
                   argstr='No argument wanted')
    iplot.read_data()
    iplot.plot_fig_1()
    iplot.plot_fig_2()
    iplot.close()

Then call the script with -h to see the command line options.

This module was written by Matthias Cuntz while at Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE), Nancy, France.

copyright:

Copyright 2020- Matthias Cuntz, see AUTHORS.rst for details.

license:

MIT License, see LICENSE for details.

The following classes are provided:

mcPlot([desc, argstr, parents])

Standard plotting class of Matthias Cuntz

History

• Written Sep 2020 by Matthias Cuntz (mc (at) macu (dot) de)

• Write standard output file into current folder, Nov 2021, Matthias Cuntz

• Change from NCL amwg color palette to pyjams amwg, May 2021, Matthias Cuntz

• Add **kwargs to plot_save, May 2022, Matthias Cuntz

• Add self.transparent to pdf output in plot_save, May 2022, Matthias Cuntz

• Add –transparent as a standard option, May 2022, Matthias Cuntz

• Add left, bottom, top to standard layout options, Jul 2022, Matthias Cuntz

• Add –dpi as a standard option, Jan 2023, Matthias Cuntz

• Use helper.filebase, Mar 2023, Matthias Cuntz

• Replace plotly with hvplot, May 2023, Matthias Cuntz

• Set filename without suffix as default plot name, Jun 2023, Matthias Cuntz

• Removed space from print of plot filename, Jun 2023, Matthias Cuntz

• Set default options if not given on command line, Jan 2024, Matthias Cuntz

• Add method print_layout_options, Mar 2024, Matthias Cuntz

• Use DejaVuSans and DejaVueSerif as standard fonts if not LaTeX because they come with matplotlib, May 2024, Matthias Cuntz

• parents argument to ArgumentParser, Jul 2024, Matthias Cuntz

class mcPlot(desc=None, argstr=None, parents=[])

Bases: object

Standard plotting class of Matthias Cuntz

Upon initialisation, the command line arguments are gathered (get_command_line_arguments), the output type is set (set_output_type), standard layout options are added to self (set_layout_options), global rcParams are set for Matplotlib (set_matplotlib_rcparams), and the output plotting file is opened (plot_begin).

desc

Description for command line parser, which will be shown when called with -h.

Type:

string, optional

argstr

String given as description for the positional arguments.

Type:

string, optional

parents

ArgumentParser or list of ArgumentParser objects whose arguments should also be included. Note, these parsers should be initialised with add_help=False. Otherwise, the ArgumentParser will see two -h/–help options and raise an error.

Type:

ArgumentParser or list, optional

get_command_line_arguments(desc=None, argstr=None, parents=[])

Standard command line parser with the default arguments such as plot type, filename, etc. If extra arguments are needed, one can pass an ArgumentParser as parent, which was initialised with add_help=False.

plot_end() or plot_stop() or plot_close() or end() or stop()

Finish, closing opened output files.

plot_save(fig, \*\*kwargs)

Save, close or show figure.

set_layout_options()

Sets the colours and styles that can be used in plots. One can either copy this routine into the extending class and adapt it, or add a new method that resets some of the layout options and call it in the initialisation of the extending class, or simply set layout options in the initialisation of the extending class.

Notes

Several more methods are defined, which should probably not be changed.

plot_begin() or plot_start()

Open output file and similar at the beginning.

plot_test()

A simple plot as an example.

print_layout_options(rcparams=False)

Prints current layout options and optionally matplotlib’’s rcParams.

set_matplotlib_rcparams()

Set rcParams of Matplotlib depending on output type, and chosen layout. rcParams can also be re-set in the initialisation of the extending class.

set_output_type()

Set the format of the output such as pdf or png.

Examples

The simplest way to use mcPlot is to extend the class:

import numpy as np
from pyjams import mcPlot

class PlotIt(mcPlot):
    def __init__(self, *args, **kwargs):
        super().__init__(*args, **kwargs)
        # change e.g. colors
        self.lcol1 = 'cyan'

    def read_data(self):
        # do something
        self.dat = np.arange(10)

    def plot_fig_1(self):
        import matplotlib.pyplot as plt

        self.ifig += 1
        fig = plt.figure(self.ifig)

        sub = fig.add_axes([0.125, 0.667, 0.3375, 0.233])

        larr = sub.plot(self.dat)
        plt.setp(larr[-1], linestyle='-', linewidth=self.lwidth,
                 marker='', color=self.lcol1)

        self.plot_save(fig)

    def plot_fig_2(self):
        import matplotlib.pyplot as plt

        self.ifig += 1
        fig = plt.figure(self.ifig)

        sub = fig.add_axes([0.125, 0.667, 0.3375, 0.233])

        larr = sub.plot(2*self.dat)
        plt.setp(larr[-1], linestyle='-', linewidth=self.lwidth,
                 marker='', color=self.lcols[-1])

        self.plot_save(fig)

if __name__ == '__main__':
    iplot = PlotIt(desc='Test mcPlot')
    iplot.read_data()
    iplot.plot_fig_1()
    iplot.plot_fig_2()
    iplot.close()

Then call the script with -h to see the command line options.

Methods

close()	Alias for plot_end()
end()	Alias for plot_end()
get_command_line_arguments([desc, argstr, ...])	Standard command line parser with default arguments such as plot type, filename, etc.
plot_begin()
plot_close()	Alias for plot_end()
plot_end()	Finish plotting
plot_save(fig, **kwargs)	Save figure into output file
plot_start()	Alias for plot_begin()
plot_stop()	Alias for plot_end()
plot_test()	A simple test plot
print_layout_options([rcparams])	Print the current layout options
set_layout_options()	Standard layout options that can be used in plotting methods
set_matplotlib_rcparams()	Set rcParams depending on output type and other options such as using LaTeX, serif fonts, etc.
set_output_type()	Set type of chosen output.

close()

Alias for plot_end()

end()

Alias for plot_end()

get_command_line_arguments(desc=None, argstr=None, parents=[])

Standard command line parser with default arguments such as plot type, filename, etc.

If extra arguments are needed, one should copy this routine into an extending class and adapt it to its needs, keeping the existing optional arguments.

Parameters:

• desc (string, optional) – Description for command line parser, which will be shown when called with -h.

• argstr (string, optional) – String given as description for the positional arguments.

• parents (ArgumentParser or list, optional) – ArgumentParser or list of ArgumentParser objects whose arguments should also be included. Note, these parsers should be initialised with add_help=False. Otherwise, the ArgumentParser will see two -h/–help options and raise an error.

Notes

Standard command line arguments are:

positional arguments:
  args                  Text will be replaced by `argstr`.

optional arguments:
  -h, --help            show this help message and exit
  -p plotname, --plotname plotname
                        Name of plot output file for types pdf,
                        html, d3, or hvplot, and name basis for type
                        png (default: calling_filename without
                        extension).
  -s, --serif           Use serif font; default sans serif.
  -t outtype, --type outtype
                        Output type is pdf, png, html, d3, or hvplot
                        (default: open screen windows).
  -u, --usetex          Use LaTeX to render text in pdf, png and
                        html.
  -w, --white           White lines on transparent or black
                        background; default: black lines on
                        transparent or white background.
  --dpi number          Dots Per inch (DPI) for non-vector output
                        types or rasterized maps in vector output
                        (default: 300).
  --transparent         Transparent figure background
                        (default: black or white).

Examples

iplot = mcPlot(desc="Test Matthias' plotting class.",
               argstr="directory file")

plot_begin()

plot_close()

Alias for plot_end()

plot_end()

Finish plotting

Close output file if appropriate or show interactive plots.

plot_save(fig, **kwargs)

Save figure into output file

Parameters:

• fig (matplotlib.figure.Figure) – Matplotlib figure object. argstr : string, optional

• kwargs (dict, optional) – Additional keywords will be passed to savefig for ‘pdf’, ‘png’, and ‘html’; to fig_to_html for ‘d3’, and to save for ‘hvplot’. This overwrites self.transparent, self.bbox_inches, self.pad_inches for ‘pdf’, ‘png’, and ‘html’.

plot_start()

Alias for plot_begin()

plot_stop()

Alias for plot_end()

plot_test()

A simple test plot

print_layout_options(rcparams=False)

Print the current layout options

Examples

Setting layout options in initialisation

pp = pyjams.mcPlot()
pp.print_layout_options(rcparams=True)

set_layout_options()

Standard layout options that can be used in plotting methods

One can either copy this routine into an extending class and adapt it, or add a new method that resets some of the layout options and call it in the initialisation of an extending class, or simply set layout options in the initialisation of an extending class.

Current layout options are:

Option	Description
self.nrow	number of rows of subplots per figure
self.ncol	number of columns of subplots per figure
self.left	left space on page
self.right	right space on page
self.bottom	lower bottom space on page
self.top	upper top space on page
self.hspace	(horizontal) x-space between subplots
self.vspace	(vertical) y-space between subplots
self.textsize	standard text size
self.dxabc	% of (max-min) shift to the right of left y-axis for a,b,c,… labels
self.dyabc	% of (max-min) shift up from lower x-axis for a,b,c,… labels
self.lwidth	line width
self.elwidth	errorbar line width
self.alwidth	axis line width
self.msize	marker size
self.mwidth	marker edge width
self.fgcolor	foreground colour
self.bgcolor	background colour
self.mcols	list of marker colours
self.mcol1	marker colour 1
self.mcol2	marker colour 2
self.mcol3	marker colour 3
self.mcol4	marker colour 4
self.mcol5	marker colour 5
self.lcol1	list of line colours
self.lcol2	line colour 1
self.lcol3	line colour 2
self.lcol4	line colour 3
self.lcol5	line colour 4
self.lcols	line colour 5
self.ldashes	list of line styles
self.llxbbox	x-anchor legend bounding box
self.llybbox	y-anchor legend bounding box
self.llrspace	spacing between rows in legend
self.llcspace	spacing between columns in legend
self.llhtextpad	pad between the legend handle and text
self.llhlength	length of the legend handles
self.frameon	if True: draw a frame around the legend. If None: use rc
self.dpi	DPI of non-vector figure output
self.transparent	True for transparent background in figure
self.bbox_inches	Bbox in inches. If ‘tight’, try to figure out the tight bbox of the figure
self.pad_inches	Amount of padding when bbox_inches is ‘tight’

Examples

Setting layout options in initialisation

class UsemcPlot(mcPlot):
    def __init__(self, *args, **kwargs):
        super().__init__(*args, **kwargs)
        self.lcol1     = 'black'
        self.mynewcol = 'red'
        # reset global values after colour changes, etc.
        self.set_matplotlib_rcparams()

set_matplotlib_rcparams()

Set rcParams depending on output type and other options such as using LaTeX, serif fonts, etc.

rcParams can be overwritten in the initialisation of an extending class.

Current parameters set are:

Parameter	Options
ps	papersize, usedistiller
figure	figsize, edgecolor, facecolor
text	usetex, latex.preamble, color
font	family, sans-serif, size
savefig	dpi, format, edgecolor, facecolor
axes	linewidth, edgecolor, facecolor, labelcolor, prop_cycle
boxplot	boxprops.color, capprops.color, flierprops.color, flierprops.markeredgecolor, whiskerprops.color
grid	color
lines	linewidth, color
patch	edgecolor
path	simplify
xtick	color
ytick	color

Examples

Setting rcParams in initialisation

class UsemcPlot(mcPlot):
    def __init__(self, *args, **kwargs):
        import matplotlib as mpl
        super().__init__(*args, **kwargs)
        mpl.rc('grid', color='red')
        mpl.rcParams['boxplot.boxprops.color'] = 'blue'

set_output_type()

Set type of chosen output.

Fall back to standard html mode if mlpd3 or hvplot modules are not installed.