Matplotlib

Matplotlib

• Matplotlib is a visualization library built on Numpy arrays

• Convention for importing Matplotlib

  import matplotlib as mpl
  import matplotlib.pyplot as plt
  
  plt.style.use('classic')

• Matplotlib was originally written as a Python alternative for MATLAB and has two interfaces:

  • A MATLAB style interface (pyplot)

  • An object oriented interface (Figure, Axes)

Displaying Plots

• Plotting from a script

  plt.show()

• Plotting from an IPython shell

  %matplotlib
  import matplotlib.pyplot as plt
  
  # use plt.draw() to force an update

Matplotlib Script Example

import matplotlib.pyplot as plt
import numpy as np

x = np.linspace(0, 10, 100)
plt.plot(x, np.sin(x))
plt.plot(x, np.cos(x))

plt.show()

Matplotlib IPython Example

%matplotlib

import matplotlib.pyplot as plt
import numpy as np

x = np.linspace(0, 10, 100)
fig = plt.figure() # create a plot window
plt.plot(x, np.sin(x))
plt.plot(x, np.cos(x))

MATLAB-style Interface Example

plt.figure() # create a plot

# create a first panel and set the axis
plt.subplot(2, 1, 1) # (rows, columns, panel #)
plt.plot(x, np.sin(x))

# create a second panel and set the axis
plt.subplot(2, 1, 2)
plt.plot(x, np.cos(x))

Object-oriented Interface Example

# create a grid of subplots
fig, ax = plt.subplots(2)
# fig is the figure
# ax is an array of Axes objects

# Call the plot method on each Axes object
ax[0].plot(x, np.sin(x))
ax[1].plot(x, np.cos(x))

Basic Plot Customization

• Plot style

• Line colors

• Line styles

• Axes limits

• Labels

Line Color

• The color parameter can be used to adjust line color

  plt.plot(x, y, color='blue') # name
  plt.plot(x, y, color='g') # color code
  plt.plot(x, y, color='0.75') # grayscale
  plt.plot(x, y, color='#FF0000') # hex
  plt.plot(x, y, color=(1.0, 0.0, 0.0)) # RBG
  plt.plot(x, y, color='papayawhip') # HTML

Line Style

• The linestyle parameter can be used to adjust line style

  plt.plot(x, y, linestyle='solid') 
  plt.plot(x, y, linestyle='dashed') 
  plt.plot(x, y, linestyle='dashdot') 
  plt.plot(x, y, linestyle='dotted') 
  
  # OR
  
  plt.plot(x, y, linestyle='-') 
  plt.plot(x, y, linestyle='--') 
  plt.plot(x, y, linestyle='-.') 
  plt.plot(x, y, linestyle=':') 

Axes Limits

• The limits of the axes can be set in various ways:

  # configure each axis individually
  plt.xlim(xmin, xmax) 
  plt.ylim(ymin, ymax) 
  
  # all at once
  plt.axis([xmin, xmax, ymin, ymax])
  
  # automatically compute limits
  plt.axis('tight')
  
  # make axis units equal
  plt.axis('equal')

Labels

• The figure can have various labels

  # add a title
  plt.title("My Awesome Plot")
  
  # label the axes
  plt.xlabel("x")
  plt.ylabel("y")
  
  # add a legend
  plt.legend()

Discrepancies Between Interfaces

• Most of the plt functions correspond to ax methods, but there are some exceptions:
  

  plt.xlabel	\(\rightarrow\)	ax.set_xlabel()
  plt.ylabel	\(\rightarrow\)	ax.set_ylabel()
  plt.xlim	\(\rightarrow\)	ax.set_xlim()
  plt.ylim	\(\rightarrow\)	ax.set_ylim()
  plt.title	\(\rightarrow\)	ax.set_title()

Scatter Plots

• The plot method can accept a marker shape

• Example

  plt.plot(x, y, 'o', color='black')
  
  # common markers:
  # 'o', '.', ',', 'x', '+', 'v',
  # '^', '<', '>', 's', 'd'

• The scatter method is specialized for scatter plots

• Example

  plt.scatter(x, y, 'o')

Another plot Example

• The plot method accepts a wide range of arguments

• Example

  plt.plot(x, y, '-p', color='gray',
           linewidth=3,
           markersize=4,
           markerfacecolor=white,
           markeredgecolor='black',
           markeredgewidth=2)

Another scatter Example

• The scatter method has additional features for scatter plots

• Example

  rng = np.random.RandomState(0)
  x = rng.randn(100)
  y = rng.randn(100)
  colors = rng.rand(100)
  sizes = 1000* rng.rand(100)
  plt.scatter(x, y, c=colors, s=sizes,
      alpha=0.3, cmap='viridis')
  plt.colorbar()

Saving Figures to a File

• The savefig method can be used to save a figure to a file

• Example

  fig = plt.figure() # create a plot window
  # add some stuff
  fig.savefig('my_figure.png')

• Supported image formats depend on which backends are installed

  # list supported file types
  fig.canvas.get_supported_filetypes()