Skip to content

python-libraries.md

Latest commit

 

History

History
133 lines (94 loc) · 3.56 KB

python-libraries.md

File metadata and controls

133 lines (94 loc) · 3.56 KB

Using libraries

If you need help at any time, put your red sticky note on the back of your laptop. When you've finished the material on the front of this page, put your green sticky note on the back of your laptop. Then, you can proceed to the extra work on the back.

Python libraries

Python is a popular language in large part because of all the libraries that are available for it, especially for scientific computing, signal processing, data analysis, machine learning, and similar tasks.

To use a Python library, you call import, e.g.

import random

and you can then call a function provided by the library as e.g.

# generate a random number between 0 and 10, inclusive
random.randint(0, 10)

You can also import a library and assign it a convenient name by which to refer to it, as e.g.

import numpy as np
np.power(2,2)

It is also possible to import all functions of a library into the namespace without having to refer to the library explicitly, e.g.

from random import *
randint(0,10)

but this is not recommended due to the possibility of naming collision.

Visualizing data with matplotlib

Matplotlib is a popular library used for visualizing data in Python. Its syntax is similar to Matlab.

Try creating a simple plot inside a Jupyter notebook:

import numpy as np
import matplotlib.pyplot as plt

X = np.linspace(-np.pi, np.pi, 256, endpoint=True)
C,S = np.cos(X), np.sin(X)

plt.plot(X,C)
plt.plot(X,S)

plt.show()

Matplotlib can also be used to create more complicated visualizations: plots that overlay data on a map, animations, etc.

\pagebreak

Useful libraries

I have listed here some Python libraries that are especially useful for ECE students. If you have extra time, choose one that is related to your interests and try out the linked example.

scikit-learn is an extremely popular machine learning library, with already-implemented versions of many popular machine learning models. A set of examples are available at:

http://scikit-learn.org/0.19/auto_examples/index.html

networkx is a library for working with data involving networks (i.e. data that is structured with edges and vertices). You can try some examples at:

https://networkx.github.io/documentation/networkx-1.11/examples/index.html

Here is an example of a research project by one of my students, using networkx:

https://witestlab.poly.edu/blog/representing-community-network-topologies-on-geni/

seaborn is a library for making nice visualizations, using matplotlib "under the hood" but with nicer default settings, and a wide range of types of plots. You can try examples at:

http://seaborn.pydata.org/examples/

scikit-image is an image processing library. You can find examples to try at:

http://scikit-image.org/docs/0.13.x/auto_examples/index.html

pandas is a great library for working with data. If you're interested in using Python for data analysis, try out some of these examples:

https://pandas.pydata.org/pandas-docs/stable/tutorials.html

Here is an examples of a research project by one of my students, using pandas:

https://witestlab.poly.edu/blog/bridging-the-empathy-gap-between-web-developers-and-users/

numpy and scipy can be used for a wide range of signal processing tasks. In this assignment, I asked students to use these libraries to take raw IQ samples from a wireless FM radio signal, demodulate it, and turn it into an audio file that can be played back on your computer. You can try it out yourself at:

http://witestlab.poly.edu/~ffund/el9043/labs/lab1