Making a plot in Python

• In this lecture, we will learn how to make a plot with Python. Matplotlib, seaborn, and plotly libraries are introduced.
• Matplotlib is the scientific plotting library in Python. It provides functions for making visualizations such as line charts, histograms, scatter plots, and so on.
• Seaborn is also widely used plotting library, which makes cooler plot.
• Plotly is gives interactive plot, which is more fancy for presentation purposes.

matplotlib

• loading the library

  import matplotlib.pyplot as plt
  

Basic line plot

• Here is the basic usage of matplotlib.

  import matplotlib.pyplot as plt
  import numpy as np
  
  x = np.linspace(-2*np.pi, 2*np.pi, 100)
  y = np.cos(x)
  
  plt.plot(x,y)
  plt.xlabel("x")
  plt.ylabel("f = cos(x)")
  plt.show()
  

• Do not forget to call show function. Otherwise, plot is made but not shown.

Scatter plot

• By changing plot to scatter, you can make a scatter plot.

  import matplotlib.pyplot as plt
  import numpy as np
  
  x = np.linspace(-2*np.pi, 2*np.pi, 100)
  y = np.cos(x)
  
  plt.scatter(x,y)
  plt.xlabel("x")
  plt.ylabel("f = cos(x)")
  plt.show()
  

Setting limits

• The range of plotting can be set as follows.

  plt.xlim([-2, 2])
  plt.ylim([0, 10])
  

Setting labels

• x- and y-axis labels are set as follows.

  plt.xlabel("xlabel")
  plt.ylabel("ylabel")
  

Setting ticks

• Ticks can be set as follows.

  plt.xticks(np.arange(0, 2+0.1, 9.5))
  

Exercise

• Plot these two data in one figure for $-2\pi \le x \le 2\pi$.
• $y = \cos(x)$
• $y = \cos(x) + 0.5(r - 0.5)$ where $r$ is the random number of $r \in [0, 1]$. answer

Figure and Axes

• In matplotlib, there are two ways to make plots.
• One is use plt.plot(), which is shown above. This is in similar way to MATLAB (actually MATplotlib means MATLAB-like plot library).
• Another way to plot is use Figure and Axes objects. Figure controls the figure part, and Axes controls the axes part of each figure.
• This is more advanced, but I recommend to use it because it enables finer control of the figure.

Example

import numpy as np
import matplotlib.pyplot as plt

x = np.linspace(0, 4*np.pi, 100)
y = np.sin(x)

fig, ax = plt.subplots()

ax.plot(x, y)
plt.show()

Setting limits

ax.set_xlim([0, 1])
ax.set_xlim([0, 1])

Setting labels

ax.set_xlabel("x")
ax.set_ylabel("y = sin(x)")
ax.set_title("sine curve")

Bar plot

import numpy as np
import matplotlib.pyplot as plt

x = np.arange(1, 6)
y = np.array([83, 32, 54, 22, 78])

fig, ax = plt.subplots()
ax.bar(x, y)

plt.show()

Histogram

import numpy as np
import matplotlib.pyplot as plt

values = 10.0 * np.random.randn(100) + 100

fig, ax = plt.subplots()
ax.hist(values, bins=10, ec="k")

plt.show()

Displaying image

import matplotlib.pyplot as plt

img = plt.imread("image.jpg")
fig, ax = plt.subplots()
ax.imshow(img)

plt.show()

Figure configuration

import numpy as np
import matplotlib.pyplot as plt

# setting font family
plt.rcParams["font.family"] = "Arial"

x = np.linspace(0, 4*np.pi, 100)
y = np.sin(x)

# controling the figure size and dpi
fig, ax = plt.subplots(figsize=(10, 6), dpi=120)

ax.plot(x, y)

# set x and y labels
ax.set_xlabel("x-axis", fontsize=16)
ax.set_ylabel("y = sin(x)", fontsize=16)

# when changing font size
## old ticklabels
xticklabel = ax.get_xticklabels()
yticklabel = ax.get_yticklabels()

## new ticklabels
ax.set_xticklabels(labels=xticklabel, fontsize=16)
ax.set_yticklabels(labels=yticklabel, fontsize=16)

# adjusting tick parameters
ax.tick_params(direction="in", length=10, width=1)

plt.tight_layout()
plt.show()

Saving figure

import numpy as np
import matplotlib.pyplot as plt

x = np.linspace(0, 4*np.pi, 100)
y = np.sin(x)

fig, ax = plt.subplots()

ax.plot(x, y)
plt.savefig("sin.png")
plt.close()

Seaborn

• Seaborn is another visualization library, which is based on matplotlib. This makes cooler plots than matplotlib.

import seaborn as sns
import matplotlib.pyplot as plt

sns.set(style="darkgrid")

# loading dataset
iris = sns.load_dataset("iris")

# scatter plot
sns.scatterplot(data=iris)
plt.show()

# histogram
sns.histplot(iris.petal_length)
plt.show()

# pair plot -- showing all the combination
sns.pairplot(data=iris)
plt.show()