Matplotlib#

What is matplotlib?#

  • Matplotlib is a plotting library for Python

    • Claim: “make the easy things easy and the hard things possible”.

  • Capable of:

    • Interactive and non-interactive plotting

    • Producing publication-quality figures

    • Can be used for schematic diagrams

  • Closely integrated with numpy

  • Use numpy functions for reading data

  • matplotlib can plot numpy arrays easily

  • See

What does it do?#

  • People often want to have a quick look at data

    • And perhaps manipulate it

  • Large amount of functionality:

    • Line charts, bar charts, scatter plots, error bars, etc..

    • Heatmaps, contours, surfaces

    • Geographical and map-based plotting

  • Can be used

    • Via a standalone script (automatiion of plotting tasks)

    • Via ipython shell

    • Within a note book

  • All methods allow you to save your work

Basic concepts#

  • Everything is assembled by Python commands

  • Create a figure with an axes area (this is the plotting area)

  • Can create one or more plots in a figure

  • Only one plot (or axes) is active at a given time

  • Use show() to display the plot

matplotlib.pyplot contains the high-level functions we need to do all the above and more

Basic plotting#

Import numpy (alias np) and matplotlib’s plotting functionality via the pyplot interface (alias plt)

import numpy as np
import matplotlib.pyplot as plt

# If using a notebook, plots can be forced to appear in the browser
# by adding the "inline" option

%matplotlib inline
import numpy as np
import matplotlib.pyplot as plt

# Create some data points for y = cos(x) using numpy

# defining x-array
xmin = 0
xmax = 10
npts = 50
x = np.linspace(xmin, xmax, npts) 

# defining y-array, based on x-array
y = np.cos(x)

plt.plot(x, y, 'gs') # 'gs': green square
plt.show()
../_images/3_matplotlib_7_0.png

Saving images to file#

  • Use, e.g., pyplot.savefig()

  • File format is determined from the filename extension you supply

  • Commonly supports: .png, .jpg, .pdf, .ps

  • Other options to control, e.g., resolution

plt.plot(x, y, 'gs')
plt.savefig("cos_gs_plot.png")
../_images/3_matplotlib_9_0.png 
# Save image to file in different formats & options
plt.plot(x, y, 'bo')
plt.savefig("cos_plot_bc.pdf")
plt.savefig("cos_plot.png", dpi=300)
plt.show()
../_images/3_matplotlib_10_0.png

Note#

matplotlib is a very large package, an has a great many objects and methods (functions). This can be confusing.

Make sure you are looking at documentation for matplotlib.pyplot

https://matplotlib.org/stable/api/pyplot_summary.html

Can help to use fully qualified names:

import matplotlib
import matplotlib.pyplot

...
matplotlib.pyplot.plot(x, y, 'rv')

to make sure you are getting the right methods.

Plot data from a file#

Step 0#

Download two associated data files random1.dat and random2.dat from our GitHub repo.

!wget https://raw.githubusercontent.com/ARCTraining/swd5_sc_py/main/book/2_matplotlib/random1.dat
!wget https://raw.githubusercontent.com/ARCTraining/swd5_sc_py/main/book/2_matplotlib/random2.dat

--2023-04-27 08:19:12--  https://raw.githubusercontent.com/ARCTraining/swd5_sc_py/main/book/2_matplotlib/random1.dat
Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.111.133, 185.199.108.133, 185.199.109.133, ...
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.111.133|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 3300 (3.2K) [text/plain]
Saving to: ‘random1.dat.1’


random1.dat.1         0%[                    ]       0  --.-KB/s               
random1.dat.1       100%[===================>]   3.22K  --.-KB/s    in 0s      

2023-04-27 08:19:12 (72.9 MB/s) - ‘random1.dat.1’ saved [3300/3300]

--2023-04-27 08:19:12--  https://raw.githubusercontent.com/ARCTraining/swd5_sc_py/main/book/2_matplotlib/random2.dat
Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.111.133, 185.199.110.133, 185.199.109.133, ...
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.111.133|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 3300 (3.2K) [text/plain]
Saving to: ‘random2.dat.1’


random2.dat.1         0%[                    ]       0  --.-KB/s               
random2.dat.1       100%[===================>]   3.22K  --.-KB/s    in 0s      

2023-04-27 08:19:12 (70.1 MB/s) - ‘random2.dat.1’ saved [3300/3300]

Step 1#

Read in the data from the files using numpy.genfromtxt(). You should have two arrays, e.g., data1 and data2. The files contain pairs of values which we will interpret as x and y coordinates. Check what these data look like (that is, check the attributes of the resulting numpy arrays).

data1 = np.genfromtxt("random1.dat")
data2 = np.genfromtxt("random2.dat")

print (data1.shape)
print (data2.shape)

(150, 2)
(150, 2)

Step 2#

Plot data1 using matplotlib to appear as red crosses (check the online documentation for pyplot.plot). You will need x-coordinates data1[:,0] and the corresponding y-coordinates

plt.plot(data1[:,0], data1[:,1], "rx")
plt.show()
../_images/3_matplotlib_18_0.png

Step 3#

Now plot data2 to appear a green circles connected by a line.

plt.plot(data2[:,0], data2[:,1], "go-")
plt.show()
../_images/3_matplotlib_20_0.png

Step 4#

  1. How do we show data1 and data2 on the same plot?

  2. Can you find out how to add labels to the axes?

  3. Can you add a legend?

Hint: you need something like plot(x, y, '+', label = "text") for the legend

plt.plot(data1[:,0], data1[:,1], 'rx', label = "dataset 1")
plt.plot(data2[:,0], data2[:,1], 'go', label = "dataset 2")
plt.xlabel("x co-ordinate")
plt.ylabel(r"$y$ co-ordinate") # add the r'text' transform the text into LaTeX text. It is useful for math.
plt.legend(loc = "upper right", bbox_to_anchor=(1.3, 1.01))
plt.show()
../_images/3_matplotlib_22_0.png

Customisation#

There are many ways to customise a plot. These may involve interaction with other matplotlib objects.

Figure size#

# Set the figure size and add a plot
# The figure size (in inches) can be specified
fig = plt.figure(figsize=(10,3))
plt.plot(x, y)
plt.show()
../_images/3_matplotlib_25_0.png

Lines#

documentation

linestyle

description

‘-’ or ‘solid’

solid line

‘–’ or ‘dashed’

dashed line

‘-.’ or ‘dashdot’

dash-dotted line

‘:’ or ‘dotted’

dotted line

‘none’, ‘None’, ‘ ‘, or ‘’

draw nothing

 
# The linewidth, and linestyle can be changed.
# Note that for standard colours, you can define with the linestyle
plt.plot(x, y, 'k-.', linewidth=4.0)

# you can specify the color name uder "color=" or "c="
# you can use "lw=" instead "linewidth="
plt.plot(x, y*2, '--', c="orange", lw=8.0)
plt.show()
../_images/3_matplotlib_27_0.png

Markers#

See markers for full description of possible arguments.

Define a marker and change sizes

# Markers and their properties can be controlled.
# Unfilled markers: '.',+','x','1' to '4','|'
plt.plot(x,y, '+', markersize=5)
plt.plot(x,2*y, 'x', ms=10)
plt.show()
../_images/3_matplotlib_30_0.png

For a filled marker, change face colour, edge colour, edge width, and size.

# Filled markers include: 'o', 's','*','d','>','^','v', 'p', 'h'
plt.plot(x, y, "D", markerfacecolor = 'g', markeredgecolor = 'r',markersize=8, markeredgewidth=2)

# you can use short for most attributes
plt.plot(x, y*2, "H", mfc = 'y', mec = 'b',ms=10, mew=5)
plt.show()
../_images/3_matplotlib_32_0.png

Combine marker and lines

plt.plot(x, y*2, "o-r", mfc = 'y', mec = 'b',ms=5, mew=1)
plt.plot(x, y*3, ":", lw=4, c="g", marker="H", mfc = 'y', mec = 'b',ms=8, mew=1)
plt.show()
../_images/3_matplotlib_34_0.png

Axes and labels#

Set x-axis and y-axis limits.

# Set axis limits
plt.plot(x, y, 'mo-')
plt.xlim((xmax*0.25, xmax*0.75))
plt.ylim((np.cos(xmin*0.25), np.cos(xmax*0.75)))
plt.show()
../_images/3_matplotlib_36_0.png

Adjust title.

# Set title placement and font properties
plt.plot(x, y, 'x')

plt.suptitle('A plot of $cos(x)$', fontsize = 20)

# Location of the title can be controled via "loc": center, left, right
#"verticalalignment": center, top, bottom, baseline
plt.title('A Placed Title', loc = 'left', verticalalignment = 'top')


plt.show()
../_images/3_matplotlib_38_0.png

Tickmarks#

Add custom tickmarks.

# Tick marks: take the default, or set explicitly
plt.plot(x, y, 'x')

# define new position for the ticks
nticks = 4
tickpos = np.linspace(xmin, xmax, nticks)

# rotate the ticks (degrees)
plt.xticks(tickpos, rotation=30)

plt.show()
../_images/3_matplotlib_40_0.png 
plt.plot(x, y, 'x')

# set arbitrary values
labels = ['A', 'B', 'C', 'D']

# choose a different colour/size
plt.xticks(tickpos, labels, c="red", fontsize=24)

([<matplotlib.axis.XTick at 0x7f4624f2a490>,
  <matplotlib.axis.XTick at 0x7f4624f2af40>,
  <matplotlib.axis.XTick at 0x7f4624ea51c0>,
  <matplotlib.axis.XTick at 0x7f4624ff07f0>],
 [Text(0.0, 0, 'A'),
  Text(3.3333333333333335, 0, 'B'),
  Text(6.666666666666667, 0, 'C'),
  Text(10.0, 0, 'D')])
../_images/3_matplotlib_41_1.png

Annotations#

# Arrows and annotations
plt.plot(x, y, 'x')
atext = 'annotate this'
arrowtip = (1.5, 0.5)
textloc=(3, 0.75)
plt.annotate(atext, xy=arrowtip, xytext=textloc,
            arrowprops=dict(facecolor='black', shrink=0.01),)
plt.show()
../_images/3_matplotlib_43_0.png

Subplots#

  • There can be multiple plots, or subplots, within a figure

  • Use subplot() to place plots on a regular grid

subplot(nrows, ncols,
        plot_number)

  • Need to control which subplot is used

    • “Current” axes is last created

    • Or use pyplot.sca(ax)

Create subplots

(fig, axes) = plt.subplots(nrows = 2, ncols = 3, figsize=(10,7))

plt.show()
../_images/3_matplotlib_46_0.png

Axes object#

Can move between subplots by keeping a reference to the axes array

(fig, axes) = plt.subplots(nrows = 2, ncols = 2)

axes[0,0].plot(x, y, 'b-')
axes[1,1].plot(x, y, 'r:')
plt.show()
../_images/3_matplotlib_48_0.png

Configure spaces for better visualisation.

# Space between subplots may be adjusted.
# subplots_adjust(left=None, bottom=None, right=None, top=None, wspace=None, hspace=None)
(fig, axes) = plt.subplots(nrows = 2, ncols = 2)
plt.subplots_adjust(wspace = 0.5, hspace = 2.0)


axes[0,1].plot(x, y, 'b-')
axes[1,0].plot(x, y, 'r-')
plt.show()
../_images/3_matplotlib_50_0.png

General subplots using subplot2grid#

  • For more control over subplot layout, use subplot2grid()

subplot2grid(shape,
             location,
             rowspan = 1,
             colspan = 1)

  • Subplots can span more than one row or column

# For example: subplot2grid(shape, loc, rowspan=1, colspan=1)
fig = plt.figure()

ax1 = plt.subplot2grid((3, 3), (0, 0))
ax2 = plt.subplot2grid((3, 3), (0, 1), colspan=2)
ax3 = plt.subplot2grid((3, 3), (1, 0), colspan=2, rowspan=2)
ax4 = plt.subplot2grid((3, 3), (1, 2), rowspan=2)

ax1.plot(x, y, 'r-')
ax2.plot(x, y, 'g-')
ax3.plot(x, y, 'b-')
ax4.plot(x, y, 'c-')
plt.show()
../_images/3_matplotlib_52_0.png

Example: Three plots#

We are now going to try to create a plot, using subplots(), which looks like:

Step 1: Data#

You will need three sets of data.

  • For the pie chart you will need to create arrays with the four percentages.

  • The two histograms are generated from data in the files uniform.dat and normal.dat respectively.

#Pie data
pie_labels = ['A', 'B', 'C', 'D']
pie_sizes = [15, 30, 35, 20]

#Histogram 1 data
!wget https://raw.githubusercontent.com/ARCTraining/swd5_sc_py/main/book/2_matplotlib/uniform.dat
data1 = np.genfromtxt('uniform.dat')

#Histogram 2 data
!wget https://raw.githubusercontent.com/ARCTraining/swd5_sc_py/main/book/2_matplotlib/normal.dat
data2 = np.genfromtxt('normal.dat')

--2023-04-27 08:19:15--  https://raw.githubusercontent.com/ARCTraining/swd5_sc_py/main/book/2_matplotlib/uniform.dat
Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.111.133, 185.199.110.133, ...
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 15000 (15K) [text/plain]
Saving to: ‘uniform.dat.1’


uniform.dat.1         0%[                    ]       0  --.-KB/s               
uniform.dat.1       100%[===================>]  14.65K  --.-KB/s    in 0s      

2023-04-27 08:19:15 (49.2 MB/s) - ‘uniform.dat.1’ saved [15000/15000]

--2023-04-27 08:19:16--  https://raw.githubusercontent.com/ARCTraining/swd5_sc_py/main/book/2_matplotlib/normal.dat
Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...
Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 15000 (15K) [text/plain]
Saving to: ‘normal.dat.1’


normal.dat.1          0%[                    ]       0  --.-KB/s               
normal.dat.1        100%[===================>]  14.65K  --.-KB/s    in 0s      

2023-04-27 08:19:16 (49.0 MB/s) - ‘normal.dat.1’ saved [15000/15000]

Step 2: Plotting frame#

You will need to create three subplots, the total size of which can be controlled by the setting the size of the figure object via

fig.set_size_inches(width, height)

Make sure the pie chart appears in the first subplot.

fig, axes = plt.subplots(3,1)
fig.set_size_inches(4.0, 8.0)
../_images/3_matplotlib_56_0.png

Step 3: Plotting & Customisation#

Check the online documentation for the pie chart to see how to produce it

http://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.pie

And check the documentation for the histogram at

http://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.hist

The pie chart colours are:

['yellowgreen', 'gold', 'lightskyblue', 'lightcoral']

plt.suptitle ("Three plots", fontsize=20)

#Pie chart
pie_colours = ['yellowgreen', 'gold', 'lightskyblue', 'lightcoral']
pie_radius = 1.5
plt.sca(axes[0])
plt.axis('equal')
plt.pie(pie_sizes, labels = pie_labels, colors = pie_colours, 
        radius = pie_radius, startangle = 90, autopct = '%1.0f%%')

#Histogram 1
plt.sca(axes[1])
plt.hist(data1, color ='m', label = 'Uniform')
plt.legend()
plt.ylabel('$P(x)$', size =16)


#Histogram 2
plt.sca(axes[2])
plt.hist(data2, color ='g', label = 'Normal')
plt.legend()
plt.ylabel('$P(x)$', size =16)
plt.xlabel('Variable $x$', size=16)

plt.show()
../_images/3_matplotlib_58_0.png

Other type of plots & settings#

Cheatsheets & Handouts#

Customisation : matplotlibrc settings#

  • Particular settings for matplotlib can be stored in a file called the matplotlibrc file

import matplotlib

matplotlib.rc_file("/path/to/my/matplotlibrc")

  • You would edit the matplotlibrc for different journal or presentation styles, for example. You could have a separate matplotlibrc for each type of style

See https://matplotlib.org/stable/tutorials/introductory/customizing.html

Settings Example

axes.labelsize  : 9.0
xtick.labelsize : 9.0
ytick.labelsize : 9.0
legend.fontsize : 9.0
font.family     : serif  
font.serif      : Computer Modern Roman

# Marker size
lines.markersize  : 3

# Use TeX to format all text (Agg, ps, pdf backends)
text.usetex : True

Summary#

  • Builds on numpy

  • Simple, interactive plotting

  • Many examples available online

  • Good enough for publication quality images

  • Can be customised for different scenarios

Advanced topic : Matplotlib frontend and backend#

Matplotlib consists of two parts, a frontend and a backend:

  • Frontend : the user facing code i.e the interface

  • Backend : does all the hard work behind-the-scenes to render the image

There are two types of backend:

  • User interface, or interactive, backends

  • Hardcopy, or non-interactive, backends to make image files

  • e.g. Agg (png), Cairo (svg), PDF (pdf), PS (eps, ps)

Check which backend is being used with

matplotlib.get_backend()

  • Switch to a different backend (before importing pyplot) with

matplotlib.use(...)
import matplot.pyplot as plt
...

For more information:

https://matplotlib.org/stable/users/explain/backends.html#backends