Python is a dynamic programming language that allows fast, platform independent development. It is a high level programming language that uses a clear and consistent syntax, aiming for both concise and readable code. In contrast to MATLAB, it is a general purpose language, not directly aimed at numerical calculations. Additional functionality is provided by packages. For scientific functionality, the SciPy library (a package set) is key. This library adds functionality for fast numerical calculations, simulation and data visualization (and more). In this example, we will specifically make use of the NumPy and matplotlib packages.

Scripting with Python

We will reuse the example from the MATLAB and LaTeX scripting page. The thought process is exactly the same so we will not repeat all steps and only detail the code.

Below is the first fragment of the Python-script displayed. We notice several steps: first, the necessary packages are loaded. Second, the data is loaded from file and the transfer function is computed. Finally, the data is saved to file.

# Example Python script: read file data, compute, make plot and export

import matplotlib.pyplot as plt
from numpy import genfromtxt
import numpy as np

data = genfromtxt('latexMatlabScriptingAllData.csv', delimiter=',', \
                  skip_header=1)
data = np.transpose(data)  # reformat for easier data extraction

# extract data to more readable vectors
f = np.array(data[0])
V1 = np.array(data[1])
V2 = np.array(data[2])
phi = np.array(data[3])

# delete deprecated data structure
del data

# compute transfer function
T = V2 / V1
phi = -phi

# save data to file (np.c_ can be replaced by np.transpose)
np.savetxt('transfer_function_py.csv', np.c_[f, T, phi], delimiter=',', \
           header="f_Hz,T,phi_degree", comments='')

If we want to plot some data, we can make use of the (already loaded) matplotlib.pyplot class. The plotting commands from plotpy are very similar (sometimes identical) to their MATLAB counterparts. Important to note are the set_tight_layout command and explicit show method call. The first method is needed to ensure everything looks good. By default, it might be possible a title or label is (partially) out of the frame. This is fixed by this call. The second command displays the actual plot. Unlike the MATLAB counterpart, figure does not open a figure window. The show method opens a window for a specific figure (fig).

# generate plot
fig = plt.figure()
fig.add_subplot(2, 1, 1)
plt.semilogx(f, 20 * np.log10(T))
plt.xlabel('frequency [Hz]')
plt.ylabel('transfer function [V/V]')
fig.add_subplot(2, 1, 2)
plt.semilogx(f, phi)
plt.xlabel('frequency [Hz]')
plt.ylabel('phase [degree]')

# fix plot sizes (otherwise, some plots and labels might overlap)
fig.set_tight_layout(True)

# display plot
plt.show()

Finally, we need to save the plot. This is much easier with Python and pyplot than MATLAB. A single-line command suffices:

# save plot to file
fig.savefig('transfer_function_figure_py.pdf')

Source code

All source code used to create this page, including measurement data, is contained in the Python-script (.py) and CSV-file (.csv).

py-filecsv-file