What is Matplotlib ? || Describe the matplotlib ||

What is Matplotlib ? || Describe the matplotlib ||

Brief Introduction:    

• While analyzing any data, instead of looking at raw data in the form of numbers, we get a much better understanding of it, if it is represented in the form of a graph.
• Matplotlib is a very popular open-source graph plotting library that is popularly used to visualize data.
• Matplotlib can be used to draw different types of graphs including, Line graph, Scatter graph, Bar chart, Histogram, pie chart, Violin chart, Contour chart, etc.
• Matplotlib can also be used to work with image and perform different operations on them.
• Graph drawn using matplotlib can either be saved then included in presentation or reports, or they can be embedded into application using GUI toolkits like TKinter, QT and GTK.
• It is possible to create static, animated and interactive data visualizations using Matplotlib library. It is possible to zoom, pan or update the interactive figures created by matplotlib.  
• Matplotlib library was create by 'John D. Hunter ' in 2003. It is written in python.

Installation and Usage

• You can install Matplotlib and check its version as follows:  

          C:\> pip install matplotlib

• Matplotlib has multiple versions. Your installed version can be checked as follows:

          import matplotlib

          print ( matplotlib.__version__)                                # printed 3.5.2 in my case

•  Instead of importing the exhaustive 'matplotlib' module, for plotting purposes 'matplotlib.pyplot' module is imported. Since plotting functions in matplotlib library expect the input data in the form of numpy arrays, most Matplotlib programs begin with following import statements:

         import matplotlib.pyplot as plt

         import numpy as np

First Graph Using Matplotlib

• The following program shows how to use Matplotlib to draw a simple graph.

         import matplotlib.pyplot as plt

         import numpy as np

         xvalues = np.linespace (0 , 2, 100 )

         yvalues = np.linespace (0 , 2,  100 )

         plt.plot ( xvalues , yvalues )

         plt.title (" Simple plot " )

         plt.xlabel ( " x values " )

         plt.ylabel ( " y values " )

         plt.show ( )

        On execution, the program produces output shown in figure 

   image 1

• The call to linespace ( )  create  100 values in the range 0 to 2. Since we wanted to draw a linear graph, we kept xvalues and yvalues same.
• Plot ( ) function is responsible for plotting the values in xvalues and yvalues  arrays and title ( ), xlabel ( ) and ylabel ( )  display the title of the graph, x-axis label and y-axis label respectively.
• Show ( ) function manages to display the plot.
• Note that once the graph is displayed, as we move the cursor within the graph, its current position in the bottom right corner of the figure.
• The other buttons in the toolbar at the bottom of the figure let you zoom or pan the figure or even save it to the disk.

Plotting Multiple Graphs

• The following program shows how to plot different growth-rates along with legend for each.

         import matplotlib as mpl

         import matplotlib.pyplot as plt

         import numpy as np

         xvalues = np.linespace (1, 5, 100 )

         ylin_values = np.linespace (1, 5, 100 )

         yquad_values = xvalues * xvalues

         ycubic_values = xvalues * xvalues * xvalues

         ylog_values = np.ylog(xvalues)

         ynlogn_values = xvalues *np.log(xvalues)

         yexp_values = 2** xvalues

         mpl.rcParams ['toolbar' ] =' None '

         plt.plot(xvalues, ylin_values, label = ' Linear ' )

         plt.plot(xvalues, yquad_values, label = ' Quadratic ' )

         plt.plot(xvalues, ycubic_values, label = ' Cubic ' )

         plt.plot(xvalues, ylog_values, label = ' Logarithmic ' )

         plt.plot(xvalues, ynlogn_values, label = ' Log Linear ' )

         plt.plot(xvalues, yexp_values, label = ' Exponential ' )

         plt.legend ( )

         plt.title ( " Rates of growth " )

         plt.xlabel (" x values " )

         plt.ylabel (y values " )

         plt.show ( )

• On execution, the program produces output shown in figure  2

image2

• Multiple graphs are drawn here for different growth rates of y values for x values present in the array xvalues.
• The legend keyword argument of plot ( ) function is used to specify the legend for each graph drawn. They get displayed when the legend ( ) function is called.
• Note that we do not have the toolbar at the bottom of the plot.  This has been achieved through the statement.

        import matplotlib as mpl

        mpl.rcParams[ 'toolbar ' ] =' None '

• Note that this statement should be executed before any plotting activity begins.

 Two Ways to use Matplotlib

• There are 2 ways to use the matplotlib library:
• Using functions in matplotlib.pyplot  module. This is the method that we have adopted in the above programs.
• Using the multiple objects present in the Matplotlib library.

• Which approach--function-based or object-based --should be used for plotting graph? Well, for simple graphs function-based approach is good enough. However, if we wish to have a finer control over each element of a graph, then we should take the object-based approach.
• In the rest of the chapter we would use the object-based approach to draw different graphs.

Using Object to Plot Graphs

• The following program shows how to draw  the plots shown in figure 2 using object in the Matplotlib library :

          import matplotlib as mpl

         import matplotlib.pyplot as plt

         import numpy as np

         xvalues = np.linespace (1, 5, 100 )

         ylin_values = np.linespace (1, 5, 100 )

         yquad_values = xvalues * xvalues

         ycubic_values = xvalues * xvalues * xvalues

         ylog_values = np.ylog(xvalues)

         ynlogn_values = xvalues *np.log(xvalues)

         yexp_values = 2** xvalues

         mpl.rcParams ['toolbar' ] =' None '

         (fig,ax) = plt.subplots ( )                         # returns a tuple

         ax.plot(xvalues, ylin_values, label = ' Linear ' )

         ax.plot(xvalues, yquad_values, label = ' Quadratic ' )

         ax.plot(xvalues, ycubic_values, label = ' Cubic ' )

         ax.plot(xvalues, ylog_values, label = ' Logarithmic ' )

         ax.plot(xvalues, ynlogn_values, label = ' Log Linear ' )

         ax.plot(xvalues, yexp_values, label = ' Exponential ' )

         ax.legend ( )

         ax.set_title ( " Rates of growth " )

         ax.set_xlabel (" x values " )

         ax.set_ylabel (y values " )

         fig.show ( )

• Here, we should first understanding the concept of Figure  and Axes.
• Figure  is an object that represents the whole figure, including the graph, title, label and legend.
• Axes represents the plotting area. So, Axes are contained with a figure. Usually, they are created together through the statement (fig,ax ) = plt.subplots ( ). It create a Figure with one Axes.
• If we wish can first create a Figure  and then add  Axes later as shown below:

        fig = plt.figure ( )       # create an empty figure no Axes

        ax =fig.subplots ( )     # add a single Axes to the FIgure

• Note that Axes represents multiple sub-plots in a figure and is not a plural of Axes that represents X-axis and y-axis.