Your first plot
Matplotlib contains a module called pyplot which enables MATLAB style plotting and customisation. We will be using this throughout this mini-course. Now, in a new terminal type the following:
import matplotlib.pyplot as plt
If you have never imported a module before, take a look here. Something slightly different about the above command is the addition of as plt - plt is used as an alias for matplotlib.pyplot (you could have called it anything but plt is often used in this case). All this does is enables you to access everything in matplotlib.pyplot by just typing plt i.e. instead of typing matplotlib.pyplot.function each time. So, all we have to now type is plt.function which is both quicker and tidier (which makes for easier reading).
Let’s start by creating some new data which we’ll do using numpy (see here if you haven’t used numpy before) - in a new terminal, type the following:
import numpy as np
data=np.array([np.arange(0,10,1), np.arange(0,30,3)])
To make things more readable, lets assign this data to two separate variables:
x=data[0,:]
y=data[1,:]
A basic line plot
To create a line plot, we will use matplotlib’s basic plot() function (remember you can find out how to use it by typing plt.plot?). We need to pass in our data - type the following to create the basic line plot:
plt.plot(x,y)
plt.show()
You can change your line style by passing in a third argument - to create a dashed line type:
plt.plot(x,y, '--')
plt.show()
You can also change the colour - for a red dashed line:
plt.plot(x,y, 'r--')
plt.show()
There are lots of available options for modifying your marker style - look here or check out the function help (plt.plot?)
A basic scatter plot
Using the same points, we can use plt.plot() to create a simple scatter plot. Unlike the line plot (the default of plt.plot()), we must specify the marker type, in the same way as we created a dashed line. For simple points, type the following:
plt.plot(x,y, 'o')
plt.show()
To make the points red:
plt.plot(x,y, 'ro')
plt.show()
To use a different symbol rather than a point (let’s go for red stars):
plt.plot(x,y, 'r*')
plt.show()
To make the points bigger, you must use the markersize option:
plt.plot(x,y, 'r*', markersize=10)
plt.show()
You’ll see that at the edge of the plot, some of the points exceed the axis and are hidden. We can adjust this by changing the plots extent by setting the x and y limits (plt.xlim()`` andplt.ylim()``` respectively):
plt.plot(x,y, 'r*', markersize=10)
plt.xlim(-1, 10)
plt.ylim(-1, 28)
plt.show()
We hardwired the axis limit values above - this works but if you will be repeatedly changing the data being used for a plot, this can end up holding you back (and hard wiring is something to avoid where possible when programming as it can end up biting you back when a variable gets hidden away as you write longer scripts). We can instead dynamically set the limits based on what we know about our x and y data and we could instead just fix an offset value - let’s say we want the axis to extend to 1 value greater than the maximum and minimum values in or x and y arrays i.e.:
offset=1
plt.plot(x,y, 'r*', markersize=10)
plt.xlim(x.min()-offset, x.max()+offset)
plt.ylim(y.min()-offset, y.max()+offset)
plt.show()
Should you now change the values in the x and y arrays, you don’t have to manually retype your extent values.