How to create Scatter Plot using Matplotlib Library

Scatter plot is one of the commonly used plots which is also thought as cousin of line plot in data visualization or data science. In this tutorial post we will discuss about scatter plot about how to create it using Matplotlib library and customize different properties of this plot based on our needs in data visualization.

In scatter plots, we represent points individually with circle, dot or any other shape instead of joining points by line segment as we mostly do in line plotting.

Let’s demonstrate some examples about how to create scatter plots. For generating scatter plots, we must start with following code to use Matplotlib library by importing it in our Jupiter notebook.

In[1]: import matplotlib.pyplot as plt  
       import numpy as np  
       %matplotlib inline

After importing matplotlib.pyplot module and numpy library in notebook, you can now easily use functions provided from these libraries as shown below;

In[2]:  x=np.linspace(0, 10, 40)   # Define x ndarray with total 20 elements b/w 0 & 10
        y=np.cos(x)                # Define y as cosine function of x
        plt.scatter(x, y)          # Generate a scatter plot

From the above plot we will notice that function y=cos(x) in y-axis is plotted against x in x-axis. But the above plot is so simple, you can customize properties of this plot as shown below;

In[3]:  plt.style.use('seaborn-whitegrid')   # Set Style of plot
        plt.scatter(x,y, c='green', marker='d', alpha=0.5) # Create scatter plot
        plt.ylim(-1.5,1.5)          # Set limits of y-axis of plot
        plt.xlabel('x')             # Set label of x-axis of plot
        plt.ylabel('cos(x)')        # Set label of y-axis of plot
        plt.title('Plotting cos(x) versus x')   # Set title of our plot

From the above lines of code In[3] you can find following points;

• we have first set style of plot from default style to ‘seaborn-whitegrid’ which is one of a list of styles provided by Matplotlib library. To list all available styles you may use command line print(plt.style.available) in IPython shell or Jupyter Notebook.

• In the next line we used scatter function to plot y against x and also we added arguments c=’green’, marker=’d’ and alpha=0.5 to this functions to customize color of point, style of point markers and blending color of point respectively.

• Using plt.ylim() function, we set maximum limits of y-axis of plot to be displayed with first argument as minimum limit and second argument as maximum limit of y-axis.

• We also set labels of x-axis and y-axis of plot using function plt.xlabel( ) and plt.ylabel( ) respectively.

• And in the final line of code we have given title of plot using function plt.title( ) with title name as string argument enclosed in it.