Lab: Plotting with matplotlib. Day 1. Copyright Oliver Serang, 2018 University of Montana Department of Computer Science

Transcription

1 1 Lab: Plotting with matplotlib Copyright Oliver Serang, 2018 University of Montana Department of Computer Science Day 1 With python, we can plot by installing matplotlib. Here is a piece of code that plots an exponential series 2 x from x = 0 to x = 9 (inclusive): # display the plot: This script will wait until the plotting window is closed. We can also change the y-axis scale to a log-scale:

2 2 # change the y-axis to log-scale pylab.yscale( log ) # display the plot: You can also use the savefig function to save figures. These figures can be.png,.pdf, or other types of files (matplotlib will automatically determine this when you give it a.png or.pdf file name!): # change the y-axis to log-scale pylab.yscale( log ) # save the plot to file my_file.png: pylab.savefig( my_file.png ) We can also use L A TEX notation to include formatted formulas in plots. The following code plots x 2 5 from x = 0 to x = 99 (inclusive) at the integers:

3 3 import math x=range(100) y=[ val**math.sqrt(2) / math.sqrt(5) for val in x ] pylab.plot(x, y, label=r $\frac{x^\sqrt{2}}{\sqrt{5}}$ ) 1. Using your memoized zombie function, z(t), from previous week s lab, create two lists: day = [0, 1, 2,...30] and zombies = [z(0), z(1), z(2),...z(30)]. Plot t against z(t), label the axes, and save the figure to a.png file using pylab.savefig function. (Include your.png file in your printed lab assignment.) 2. Replot and resave the answer to your previous question by using a log-scale on the y-axis. (Include your.png file in your printed lab assignment.) 3. Is the zombie population growing faster than the exponential function 2 t? Add another series (2 t ) to your plot and annotate these on the plot legend. (Include your.png file in your printed lab assignment.) Day 2 You can change the axis annotations or use qualitative variables using the pylab.xticks. This is shown in the following bar plot script: day_number = range(7) temperature = [ 7, 10, -3, 5, 12, 7, 3 ] # make a bar plot: pylab.bar(day_number, temperature)

4 4 # use qualitative variables: pylab.xticks(day_number, [ Monday, Tuesday, Wednesday, Thursday, Friday, Saturday, Sunday ]) # show the data Overlapping series can be difficult to see, because one will obscure the other. For this reason, we can enable transparency with alpha compositing : day_number = range(7) temperature = [ 7, 10, -3, 5, 12, 7, 3 ] temperature_2 = [ 2, 12, 4, 3, 12, 6, -2 ] # make two bar plots: pylab.bar(day_number, temperature, label= First series, alpha=0.7) pylab.bar(day_number, temperature_2, label= Another series, alpha=0.7) # use qualitative variables: pylab.xticks(day_number, [ Monday, Tuesday, Wednesday, Thursday, Friday, Saturday, Sunday ]) # draw the legend # show the data We can also control the position of the legend: pylab.legend(loc= upper right ) will draw the legend in the upper right corner of the plot. This can be useful when the automatic position of the legend would obscure much of the data plotted. 1. Using your Monte Carlo assignment that estimated π, make two lists: the first list will contain the number of iterations so far and the second list will contain the Monte Carlo estimate thus far. Plot these two lists using the number of iterations from 1 to 500 (inclusive). On the same

5 figure, plot the floating point estimate of π (using numpy.pi) using a list containing many copies of the same value: [π, π, π,... π]. Use the L A TEX notation \pi to label the y-axis with Estimate of π. (Include your.png file in your printed lab assignment.) 2. Retrieve the snowfall data (by month) for Missoula and for Bozeman. Create bar plots with labeled axes (use qualitative labels for the names of the months). (Include your.png file in your printed lab assignment.) 3. Add another bar series for Bozeman s snowfall to the same plot. Use transparency with alpha compositing so that the overlap can be distinguished. (Include your.png file in your printed lab assignment.) 5