Introduction

Comparison With Matlab

Advantages
Extensions with Pyrex
Communication

Numerical Computing in Python

A Guide for Matlab Users

B. Blais

Department of Science and Technology

Bryant University

Faculty Development Seminar - Spring 2007

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Advantages
Extensions with Pyrex
Communication

Abstract
Matlab is a commercial program used extensively in the scientific and business communities. There
are many reasons why it is very popular, including its interactive structure, clean syntax, and ability
to interface with fast compiled languages, like C. It also has many routines for signal and image
processing, optimization, and visualization.

Python is a modern language used extensively by Google and NASA, as well as many others. Like
Matlab, it also has an interactive structure, clean syntax, and the ability to interface with fast
compiled languages, like C. There are modules in Python for doing numerical work and
visualization, and thus one can make a Python-based computational environment with much the
same feel as Matlab. Python is also free, is far more versatile, and can be used in many more
applications than Matlab, including robotics, web frameworks, text processing, and others. It is
particularly good as a first language, and I have found it personally very useful in my classes.

This Faculty Development Seminar uses a “how-to” approach to setting up Python as a

computational environment, geared towards current users of Matlab or similar environments. It
explores specific applications of numerical computing, and highlights the power of using Python
both in research and in teaching. The seminar will explore my own experiences of the past year,
converting from a die-hard Matlab fan to a Python enthusiast.

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Advantages
Extensions with Pyrex
Communication

Outline

1 Introduction
2 Comparison with Matlab
3 Advantages
4 Extensions with Pyrex
5 Communication
6 Conclusions

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Where am I Coming From?

1980-1988: The BASIC Years

1989-1993: The Pascal Years (with a little Fortran)
1994-1996: The C/C++ Years
1995-2006: The Matlab Years (with C for cmex)
2003-2006: The Disenchantment Years
2006-present: The Python Year(s)

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Where am I Coming From?

1980-1988: The BASIC Years

1989-1993: The Pascal Years (with a little Fortran)
1994-1996: The C/C++ Years
1995-2006: The Matlab Years (with C for cmex)
2003-2006: The Disenchantment Years
2006-present: The Python Year(s)

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Where am I Coming From?

1980-1988: The BASIC Years

1989-1993: The Pascal Years (with a little Fortran)
1994-1996: The C/C++ Years
1995-2006: The Matlab Years (with C for cmex)
2003-2006: The Disenchantment Years
2006-present: The Python Year(s)

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Where am I Coming From?

1980-1988: The BASIC Years

1989-1993: The Pascal Years (with a little Fortran)
1994-1996: The C/C++ Years
1995-2006: The Matlab Years (with C for cmex)
2003-2006: The Disenchantment Years
2006-present: The Python Year(s)

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Where am I Coming From?

1980-1988: The BASIC Years

1989-1993: The Pascal Years (with a little Fortran)
1994-1996: The C/C++ Years
1995-2006: The Matlab Years (with C for cmex)
2003-2006: The Disenchantment Years
2006-present: The Python Year(s)

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Where am I Coming From?

1980-1988: The BASIC Years

1989-1993: The Pascal Years (with a little Fortran)
1994-1996: The C/C++ Years
1995-2006: The Matlab Years (with C for cmex)
2003-2006: The Disenchantment Years
2006-present: The Python Year(s)

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Where am I Coming From?

1980-1988: The BASIC Years

1989-1993: The Pascal Years (with a little Fortran)
1994-1996: The C/C++ Years
1995-2006: The Matlab Years (with C for cmex)
2003-2006: The Disenchantment Years
2006-present: The Python Year(s)

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

What is Python?

Flexible, powerful language

Multiple programming paradigms
Easy, clean syntax
Large community of support
“Batteries included”
Free as in “free beer”
Free as in “free speech”

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Non-Numerical Projects I’ve Done with Python

Making Aemilia’s and Aoife’s web-page

web.bryant.edu/˜bblais/gallery
Curriculum Committee web-page
Football Statistics
Student Picture Game
Posting Grades
Robot Simulator
Robot Programming Language

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Numerical Projects I’ve Done with Python

Neural Simulators
Plasticity: rate-based
Splash: spike-based
Physics Projects
Simulating falling objects
Simulating flipping coins
Signal processing for SETI
AI and Robotics Projects
Analysis of finance data
Voice recognition
Mechanisms of the Mind Projects:
Supervised and Unsupervised Learning
Associative Networks
Bayesian Statistical Inference Notes
B. Blais Numerical Computing in Python
 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Example Python Code

from math import sin,pi

def sinc(x):
’’’Compute the sinc function:
sin(pi*x)/(pi*x)’’’

try:
val = (x*pi)
return sin(val)/val
except ZeroDivisionError:
return 1.0

input=[0,0.1,0.5,1.0] # list of input values

output=[sinc(x) for x in input]

print output

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Packages for a Useful Computational Environment

Minimum
python - the base language
numpy - array class, numerical routines
scipy - higher level scientific routines (de-
pends on numpy)
matplotlib - visualization
ipython - a more flexible python shell

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Packages for a Useful Computational Environment

Minimum
python - the base language
numpy - array class, numerical routines
scipy - higher level scientific routines (de-
pends on numpy)
matplotlib - visualization
ipython - a more flexible python shell

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Packages for a Useful Computational Environment

Minimum
python - the base language
numpy - array class, numerical routines
scipy - higher level scientific routines (de-
pends on numpy)
matplotlib - visualization
ipython - a more flexible python shell

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Packages for a Useful Computational Environment

Minimum
python - the base language
numpy - array class, numerical routines
scipy - higher level scientific routines (de-
pends on numpy)
matplotlib - visualization
ipython - a more flexible python shell

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Packages for a Useful Computational Environment

Minimum
python - the base language
numpy - array class, numerical routines
scipy - higher level scientific routines (de-
pends on numpy)
matplotlib - visualization
ipython - a more flexible python shell

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Packages for a Useful Computational Environment

Useful
pyrex - writing fast compiled extensions
(like cmex, but way better)
wxpython - GUI library
pywin32 - Windows COM Interface
BeautifulSoup - HTML Parser
xlrd, pyXLWriter - Reading/Writing Excel Spread-
sheets

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Packages for a Useful Computational Environment

Useful
pyrex - writing fast compiled extensions
(like cmex, but way better)
wxpython - GUI library
pywin32 - Windows COM Interface
BeautifulSoup - HTML Parser
xlrd, pyXLWriter - Reading/Writing Excel Spread-
sheets

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Packages for a Useful Computational Environment

Useful
pyrex - writing fast compiled extensions
(like cmex, but way better)
wxpython - GUI library
pywin32 - Windows COM Interface
BeautifulSoup - HTML Parser
xlrd, pyXLWriter - Reading/Writing Excel Spread-
sheets

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Packages for a Useful Computational Environment

Useful
pyrex - writing fast compiled extensions
(like cmex, but way better)
wxpython - GUI library
pywin32 - Windows COM Interface
BeautifulSoup - HTML Parser
xlrd, pyXLWriter - Reading/Writing Excel Spread-
sheets

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab What is Python?
Advantages Projects
Extensions with Pyrex What You Need
Communication

Packages for a Useful Computational Environment

Useful
pyrex - writing fast compiled extensions
(like cmex, but way better)
wxpython - GUI library
pywin32 - Windows COM Interface
BeautifulSoup - HTML Parser
xlrd, pyXLWriter - Reading/Writing Excel Spread-
sheets

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Functions: Data types and Files

Matlab Code: f2c.m and c2f.m
function c=f2c(f)
c=(f-32)*(100/180);

function f=c2f(c)
f=(180/100)*c+32;

Python Code: convert.py

def f2c(f):
return (f-32)*(100.0/180.0)

def c2f(c):
return (180.0/100.0)*c+32
B. Blais Numerical Computing in Python
 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Interactive Environment

Running the Matlab Code

>> a=f2c(212)

a =

100

>> b=c2f(-40)

b =

-40

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Interactive Environment

Running the Python Code: Namespaces

In [1]:from convert import *

In [2]:a=f2c(212)

In [3]:a
Out[3]:100.0

In [4]:b=c2f(-40)

In [5]:b
Out[5]:-40.0

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Interactive Environment

Object-oriented or Procedural
In [7]:import convert

In [8]:a=convert.f2c(212)

In [9]:a
Out[9]:100.0

In [10]:dir(convert)
Out[10]:[’__builtins__’, ’__doc__’,
’__file__’, ’__name__’,
’c2f’, ’f2c’]

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Interactive Environment
Object-oriented or Procedural
In [1]:from Temperature import *
In [2]:t=Temperature(f=32)
In [3]:print t.c
0.0

In [4]:print t.k
273.15

In [5]:t.c=-40

In [6]:t.k
Out[6]:233.14999999999998

In [7]:t.f
Out[7]:-40.0

In [8]:t.k=350

In [9]:t.c
Out[9]:76.850000000000023

In [10]:t.f
Out[10]:170.33000000000004

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Interactive Environment
Object-oriented or Procedural
class Temperature(object):
coefficients = {’c’: (1.0, 0.0, -273.15), ’f’: (1.8, -273.15, 32.0)}
def __init__(self, **kwargs):
try:
name, value = kwargs.popitem( )
except KeyError:
name, value = ’k’, 0
setattr(self, name, float(value))
def __getattr__(self, name):
try:
eq = self.coefficients[name]
except KeyError:
raise AttributeError, name
return (self.k + eq[1]) * eq[0] + eq[2]
def __setattr__(self, name, value):
if name in self.coefficients:
# name is c or f -- compute and set k
eq = self.coefficients[name]
self.k = (value - eq[2]) / eq[0] - eq[1]
elif name == ’k’:
object.__setattr__(self, name, value)
else:
raise AttributeError, name

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Interactive Environment
Object-oriented or Procedural
In [1]:from Temperature import *
In [2]:t=Temperature(f=32)
In [3]:print t.c
0.0

In [4]:print t.k
273.15

In [5]:t.c=-40

In [6]:t.k
Out[6]:233.14999999999998

In [7]:t.f
Out[7]:-40.0

In [8]:t.k=350

In [9]:t.c
Out[9]:76.850000000000023

In [10]:t.f
Out[10]:170.33000000000004

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Getting Help

In Matlab, help is contained in the file

>> help fft

FFT Discrete Fourier transform.

FFT(X) is the discrete Fourier transform (DFT) of vector X. For
matrices, the FFT operation is applied to each column. For N-D
arrays, the FFT operation operates on the first non-singleton
dimension.

...

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Getting Help

In Python, help is contained in the object

In [14]:help(fft)
Help on function fft in module numpy.fft.fftpack:

fft(a, n=None, axis=-1)

fft(a, n=None, axis=-1)

Return the n point discrete Fourier transform of a. n defaults to

the length of a. If n is larger than the length of a, then a will
be zero-padded to make up the difference. If n is smaller than
the length of a, only the first n items in a will be used.

...

. . . and everything is an object: lists, arrays, functions, integers,

etc. . .

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Getting Help

In Python, help is contained in the object

In [14]:help(fft)
Help on function fft in module numpy.fft.fftpack:

fft(a, n=None, axis=-1)

fft(a, n=None, axis=-1)

Return the n point discrete Fourier transform of a. n defaults to

the length of a. If n is larger than the length of a, then a will
be zero-padded to make up the difference. If n is smaller than
the length of a, only the first n items in a will be used.

...

. . . and everything is an object: lists, arrays, functions, integers,

etc. . .

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Getting Help
Namespace is important for getting Help
In [19]:import scipy

In [20]:help(scipy)

Help on package scipy:

NAME
scipy

FILE
/usr/local/lib/python2.5/site-packages/scipy/__init__.py

DESCRIPTION
SciPy --- A scientific computing package for Python
===================================================
...
Available subpackages
---------------------
ndimage --- n-dimensional image package [*]
stats --- Statistical Functions [*]
signal --- Signal Processing Tools [*]
...

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Help

In Python, all assignments are name assignments

In [15]:d=fft # assign a new name

In [16]:help(d)
Help on function fft in module numpy.fft.fftpack:

fft(a, n=None, axis=-1)

fft(a, n=None, axis=-1)

Return the n point discrete Fourier transform of a. n defaults to

the length of a. If n is larger than the length of a, then a will
be zero-padded to make up the difference. If n is smaller than
the length of a, only the first n items in a will be used.

. . . so all parameters are pass by reference. We’re all adults

here.

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Help

In Python, all assignments are name assignments

In [15]:d=fft # assign a new name

In [16]:help(d)
Help on function fft in module numpy.fft.fftpack:

fft(a, n=None, axis=-1)

fft(a, n=None, axis=-1)

Return the n point discrete Fourier transform of a. n defaults to

the length of a. If n is larger than the length of a, then a will
be zero-padded to make up the difference. If n is smaller than
the length of a, only the first n items in a will be used.

. . . so all parameters are pass by reference. We’re all adults

here.

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Some Useful Help Tips in IPython

Tab completion, especially for a methods list

In [4]:numpy.d<TAB>
numpy.delete numpy.diagflat numpy.digitize numpy.dot numpy.dstack
numpy.deprecate numpy.diagonal numpy.disp numpy.double numpy.dtype
numpy.diag numpy.diff numpy.divide numpy.dsplit

‘?’ notation for help

In [4]:numpy.diag?
Type: function
Base Class: <type ’function’>
String Form: <function diag at 0xb618e8b4>
Namespace: Interactive
File: /usr/local/lib/python2.5/site-packages/numpy/lib/twodim_base.py
Definition: numpy.diag(v, k=0)
Docstring:
returns a copy of the the k-th diagonal if v is a 2-d array
or returns a 2-d array with v as the k-th diagonal if v is a
1-d array.

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Some Useful Help Tips in IPython

Tab completion, especially for a methods list

In [4]:numpy.d<TAB>
numpy.delete numpy.diagflat numpy.digitize numpy.dot numpy.dstack
numpy.deprecate numpy.diagonal numpy.disp numpy.double numpy.dtype
numpy.diag numpy.diff numpy.divide numpy.dsplit

‘?’ notation for help

In [4]:numpy.diag?
Type: function
Base Class: <type ’function’>
String Form: <function diag at 0xb618e8b4>
Namespace: Interactive
File: /usr/local/lib/python2.5/site-packages/numpy/lib/twodim_base.py
Definition: numpy.diag(v, k=0)
Docstring:
returns a copy of the the k-th diagonal if v is a 2-d array
or returns a 2-d array with v as the k-th diagonal if v is a
1-d array.

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Zen

In [23]:import this
The Zen of Python, by Tim Peters

Beautiful is better than ugly.

Explicit is better than implicit.
Simple is better than complex.
Complex is better than complicated.
Flat is better than nested.
Sparse is better than dense.
Readability counts.
Special cases aren’t special enough to break the rules.
Although practicality beats purity.
Errors should never pass silently.
Unless explicitly silenced.
In the face of ambiguity, refuse the temptation to guess.
There should be one-- and preferably only one --obvious way to do it.
Although that way may not be obvious at first unless you’re Dutch.
Now is better than never.
Although never is often better than *right* now.
If the implementation is hard to explain, it’s a bad idea.
If the implementation is easy to explain, it may be a good idea.
Namespaces are one honking great idea -- let’s do more of those!

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Golden Ratio

Matlab Python
function goldfract(n) def goldfract(N):
%GOLDFRACT Golden ratio continued """GOLDFRACT(N)
% fraction. Golden ratio continued fraction
% GOLDFRACT(n) displays n terms. Displays N terms."""
p = ’1’; p = ’1.0’
for k = 1:n for k in range(N):
p = [’1+1/(’ p ’)’]; p = ’1.0+1.0/(’ +p+ ’)’
end print p
p
p = 1; p = 1
q = 1; q = 1
for k = 1:n for k in range(N):
s = p; s = p
p = p + q; p = p + q
q = s; q = s
end
print ’%d/%d’ % (p,q)
p = sprintf(’%d/%d’,p,q) p=’%f/%f’ % (p,q) # use floats
format long p=eval(p)
p = eval(p) print "%.20f" % p
format short err = (1+sqrt(5))/2 - p
err = (1+sqrt(5))/2 - p print err

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Fibonacci

Matlab Python
function f = fibonacci(n) def fibonacci(n):
% FIBONACCI Fibonacci sequence """FIBONACCI Fibonacci sequence
% f = FIBONACCI(n) generates the """
% first n Fibonacci numbers. from numpy import zeros

f=zeros(n)
f = zeros(n,1); f[0] = 1
f(1) = 1; f[1] = 2
f(2) = 2; for k in range(2,n):
for k = 3:n f[k]=f[k-1]+f[k-2]
f(k) = f(k-1) + f(k-2);
end return f

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Fibonacci

Python: Array
def fibonacci(n): Python: List
"""FIBONACCI Fibonacci sequence
def fibonacci2(n):
"""
"""FIBONACCI Fibonacci sequence
from numpy import zeros
"""
f=zeros(n)
f=[1,2] # use a list
f[0] = 1
for k in range(2,n):
f[1] = 2
f.append(f[k-1]+f[k-2])
for k in range(2,n):
f[k]=f[k-1]+f[k-2]
return f
return f

Lists are a little like cell arrays, but more flexible

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Financial Data
Get the Data
import scipy
import os
import urllib
import datetime

# get the data

start=[1,1,2007]
end=[4,25,2007]

fname=’my_yahoo_data.csv’

if not os.path.exists(fname):
url=’http://ichart.finance.yahoo.com/table.csv?s=%%5EIXIC&d=%d&e=%d&f=%d&g=d&a=%d&b=%d&c=%
print url

f = urllib.urlopen(url)

k=open(fname,"wt")
st=f.read()
k.write(st)
k.close()
f.close()

Date,Open,High,Low,Close,Volume,AdjB.Close
Blais Numerical Computing in Python
 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Financial Data
Date,Open,High,Low,Close,Volume,Adj Close
2007-04-25,2533.54,2551.39,2523.84,2547.89,2644120000,2547.89
2007-04-24,2528.39,2529.48,2509.26,2524.54,2220610000,2524.54
2007-04-23,2525.77,2531.40,2518.47,2523.67,1928530000,2523.67
...

Parse the Data

# read the data
count=0
vals=[]; dates=[]
for line in open(fname):
count=count+1

if count==1: # skip the first line

continue

val=float(line.split(’,’)[-1])
vals.append(val) # last value

date=line.split(’,’)[0].split(’-’)
dint=[int(x) for x in date] # convert to ints
dateval=datetime.date(dint[0],dint[1],dint[2]).toordinal()
dates.append(dateval) # first value

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Financial Data
Plot the Data
clf()
# plot the data
plot_date(dates,vals,’-o’)

p=scipy.polyfit(dates,vals,1)
x=arange(min(dates),max(dates),1)
y=p[0]*x+p[1]
plot(x,y,’r--’,linewidth=3)

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Rosenbrock Function of N Variables

N−1
X
2
f (x) = 100(xi − xi−1 )2 + (1 − xi−1 )2
i=1
Minimum at x0 = x1 = . . . = 1
Perform the Optimization
from scipy.optimize import fmin
def rosen(x): # The Rosenbrock function
return sum(100.0*(x[1:]-x[:-1]**2.0)**2.0 + (1-x[:-1])**2.0)

x0 = [1.3, 0.7, 0.8, 1.9, 1.2]

xopt = fmin(rosen, x0) # Nelder-Mead simplex algorithm

Optimization terminated successfully.

Current function value: 0.000066
Iterations: 141
Function evaluations: 243
[ 0.99910115 0.99820923 0.99646346 0.99297555 0.98600385]

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Rosenbrock Function of N Variables

N−1
X
2
f (x) = 100(xi − xi−1 )2 + (1 − xi−1 )2
i=1
Minimum at x0 = x1 = . . . = 1
Perform the Optimization
from scipy.optimize import fmin
def rosen(x): # The Rosenbrock function
return sum(100.0*(x[1:]-x[:-1]**2.0)**2.0 + (1-x[:-1])**2.0)

x0 = [1.3, 0.7, 0.8, 1.9, 1.2]

xopt = fmin(rosen, x0) # Nelder-Mead simplex algorithm

Optimization terminated successfully.

Current function value: 0.000066
Iterations: 141
Function evaluations: 243
[ 0.99910115 0.99820923 0.99646346 0.99297555 0.98600385]

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Rosenbrock Function of N Variables

N−1
X
2
f (x) = 100(xi − xi−1 )2 + (1 − xi−1 )2
i=1
Minimum at x0 = x1 = . . . = 1
Perform the Optimization
from scipy.optimize import fmin
def rosen(x): # The Rosenbrock function
return sum(100.0*(x[1:]-x[:-1]**2.0)**2.0 + (1-x[:-1])**2.0)

x0 = [1.3, 0.7, 0.8, 1.9, 1.2]

xopt = fmin(rosen, x0) # Nelder-Mead simplex algorithm

Optimization terminated successfully.

Current function value: 0.000066
Iterations: 141
Function evaluations: 243
[ 0.99910115 0.99820923 0.99646346 0.99297555 0.98600385]

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Fitting a Sine Wave

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Fitting a Sine Wave

Generate the Data

from pylab import *
from numpy import *
import scipy
from scipy import optimize

from bigfonts import bigfonts

bigfonts()

x=linspace(0,6e-2,100)
A,k,theta = 10, 1.0/3e-2, pi/6
y_true = A*sin(2*pi*k*x+theta)
y_meas = y_true + 2*randn(len(x))

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Fitting a Sine Wave

Fit the Data

def residuals(p, y, x):
A,k,theta = p
err = y-A*sin(2*pi*k*x+theta)
return err

def peval(x, p):

return p[0]*sin(2*pi*p[1]*x+p[2])

p0 = [20, 40, 10]

print "Initial values:",p0

plsq = optimize.leastsq(residuals, p0, args=(y_meas, x))

print "Final estimates:",plsq[0]

print "Actual values:", [A, k, theta]

B. Blais Numerical Computing in Python

 Initial Comparison
Introduction
Getting Help
Comparison With Matlab
Golden Ratio
Advantages
Fibonacci
Extensions with Pyrex
Finance
Communication
Optimization and Least Squares

Fitting a Sine Wave

Output from Program
Initial values: [20, 40, 10]
Final estimates: [-10.41111011 33.09546027 10.00631967]
Actual values: [10, 33.333333333333336, 0.52359877559829882]

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Matlab

String as Argument when given No Parentheses

Matlab Python
help spam import spam
help(spam)

Clean Syntax for Inputing Matrices

Matlab Python
a=[1 2 3 ; 4 5 6] from numpy import *
a=mat(’[1 2 3 ; 4 5 6]’)
a=matrix([ [1,2,3] , [4,5,6] ])

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Matlab

String as Argument when given No Parentheses

Matlab Python
help spam import spam
help(spam)

Clean Syntax for Inputing Matrices

Matlab Python
a=[1 2 3 ; 4 5 6] from numpy import *
a=mat(’[1 2 3 ; 4 5 6]’)
a=matrix([ [1,2,3] , [4,5,6] ])

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Matlab

String as Argument when given No Parentheses

Matlab Python
help spam import spam
help(spam)

Clean Syntax for Inputing Matrices

Matlab Python
a=[1 2 3 ; 4 5 6] from numpy import *
a=mat(’[1 2 3 ; 4 5 6]’)
a=matrix([ [1,2,3] , [4,5,6] ])

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Matlab

Clean Syntax for Inputing Range

Matlab Python
a=1:10 from numpy import *
a=r_[1:11] # 1 minus last number
b=linspace(1,10,10)
b=linspace(1,10,10) # better way

Calling a user-created script

Matlab Python
% run my script with some commands # in ipython
myscript; run myscript.py

# in regular python shell

execfile(’myscript.py’)

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Matlab

Clean Syntax for Inputing Range

Matlab Python
a=1:10 from numpy import *
a=r_[1:11] # 1 minus last number
b=linspace(1,10,10)
b=linspace(1,10,10) # better way

Calling a user-created script

Matlab Python
% run my script with some commands # in ipython
myscript; run myscript.py

# in regular python shell

execfile(’myscript.py’)

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Matlab

Better integrated plot commands

Matlab Python
x=-10:10 from pylab import *
y=x.^2 from numpy import *
plot(x,y,’-o’)
x=linspace(-10,10,20)
y=x**2

plot(x,y,’-o’)
show()

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Matlab

the “.” operators on matrices

Matlab Python
a=[1 2 3; 4 5 6; 7 8 9] from numpy import *
b=[10 20 30; 40 50 60; 70 80 90] # two choices: matrix or array class

% matrix multiply a=mat(’[1 2 3; 4 5 6; 7 8 9]’)

c=a*b b=mat(’[10 20 30; 40 50 60; 70 80 90]’)

% element-by-element multiply # matrix multiply

d=a.*b c=a*b
# element-by-element multiply on matrix
d=multiply(a,b)

a=array(a)
b=array(b)

# matrix multiply on arrays

c=dot(a,b)
# element-by-element multiply on array
d=a*b

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Namespaces: Scales to Larger Projects

Matlab Python
a=sqrt(2) % built-in import math
import mymath
% uses first fmin in path
fmin(’cos’,3,4) a=math.sqrt(2)
b=mymath.sqrt(2)

from scipy.optimize import fmin

from myopt import fmin as fmin2

from math import cos

fmin(cos,3,4) # uses scipy fmin
fmin2(cos,3,4) # uses my fmin

Free as in “beer” and “speech”

Real object-oriented programming
Can define functions in a script
B. Blais Numerical Computing in Python
 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Namespaces: Scales to Larger Projects

Matlab Python
a=sqrt(2) % built-in import math
import mymath
% uses first fmin in path
fmin(’cos’,3,4) a=math.sqrt(2)
b=mymath.sqrt(2)

from scipy.optimize import fmin

from myopt import fmin as fmin2

from math import cos

fmin(cos,3,4) # uses scipy fmin
fmin2(cos,3,4) # uses my fmin

Free as in “beer” and “speech”

Real object-oriented programming
Can define functions in a script
B. Blais Numerical Computing in Python
 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Namespaces: Scales to Larger Projects

Matlab Python
a=sqrt(2) % built-in import math
import mymath
% uses first fmin in path
fmin(’cos’,3,4) a=math.sqrt(2)
b=mymath.sqrt(2)

from scipy.optimize import fmin

from myopt import fmin as fmin2

from math import cos

fmin(cos,3,4) # uses scipy fmin
fmin2(cos,3,4) # uses my fmin

Free as in “beer” and “speech”

Real object-oriented programming
Can define functions in a script
B. Blais Numerical Computing in Python
 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Namespaces: Scales to Larger Projects

Matlab Python
a=sqrt(2) % built-in import math
import mymath
% uses first fmin in path
fmin(’cos’,3,4) a=math.sqrt(2)
b=mymath.sqrt(2)

from scipy.optimize import fmin

from myopt import fmin as fmin2

from math import cos

fmin(cos,3,4) # uses scipy fmin
fmin2(cos,3,4) # uses my fmin

Free as in “beer” and “speech”

Real object-oriented programming
Can define functions in a script
B. Blais Numerical Computing in Python
 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library of Modules for Multiple Purposes
Standard Library
String Services
string: Common string operations
re: expression operations
struct: Interpret strings as packed binary data
difflib: Helpers for computing deltas
StringIO: Read and write strings as files
cStringIO: Faster version of StringIO
textwrap: Text wrapping and filling
codecs: Codec registry and base classes
unicodedata: Unicode Database
stringprep: Internet String Preparation
fpformat: Floating point conversions

Data Types
datetime: Basic date and time types
calendar: General calendar-related functions
collections: High-performance container datatypes
heapq: Heap queue algorithm
bisect: Array bisection algorithm
array: Efficient arrays of numeric values

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library of Modules for Multiple Purposes
Standard Library
String Services
string: Common string operations
re: expression operations
struct: Interpret strings as packed binary data
difflib: Helpers for computing deltas
StringIO: Read and write strings as files
cStringIO: Faster version of StringIO
textwrap: Text wrapping and filling
codecs: Codec registry and base classes
unicodedata: Unicode Database
stringprep: Internet String Preparation
fpformat: Floating point conversions

Data Types
datetime: Basic date and time types
calendar: General calendar-related functions
collections: High-performance container datatypes
heapq: Heap queue algorithm
bisect: Array bisection algorithm
array: Efficient arrays of numeric values

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library (continued)

sets: Unordered collections of unique elements
sched: Event scheduler
mutex: Mutual exclusion support
Queue: A synchronized queue class
weakref: Weak references
UserDict: Class wrapper for dictionary objects
UserList: Class wrapper for list objects
UserString: Class wrapper for string objects
types: Names for built-in types
new: Creation of runtime internal objects
copy: Shallow and deep copy operations
pprint: Data pretty printer
repr: Alternate repr() implementation

Numeric and Mathematical Modules

math: Mathematical functions
cmath: Mathematical functions for complex numbers
decimal: Decimal floating point arithmetic
random: Generate pseudo-random numbers
itertools: Functions creating iterators for efficient looping

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library (continued)

functools: Higher order functions and operations on callable objects.
operator: Standard operators as functions.

Internet Data Handling

email: An email and MIME handling package
mailcap: Mailcap file handling.
mailbox: Manipulate mailboxes in various formats
mhlib: Access to MH mailboxes
mimetools: Tools for parsing MIME messages
mimetypes: Map filenames to MIME types
MimeWriter: Generic MIME file writer
mimify: MIME processing of mail messages
multifile: Support for files containing distinct parts
rfc822: Parse RFC 2822 mail headers
base64: RFC 3548: Base16, Base32, Base64 Data Encodings
binhex: Encode and decode binhex4 files
binascii: Convert between binary and ASCII
quopri: Encode and decode MIME quoted-printable data
uu: and decode uuencode files

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library (continued)
Structured Markup Processing Tools
HTMLParser: Simple HTML and XHTML parser
sgmllib: Simple SGML parser
htmllib: A parser for HTML documents
htmlentitydefs: Definitions of HTML general entities
xml.parsers.expat: Fast XML parsing using Expat
xml.dom: The Document Object Model API
xml.dom.minidom: Lightweight DOM implementation
xml.dom.pulldom: Support for building partial DOM trees
xml.sax: Support for SAX2 parsers
xml.sax.handler: Base classes for SAX handlers
xml.sax.saxutils: SAX Utilities
xml.sax.xmlreader: Interface for XML parsers
xml.etree.ElementTree: The ElementTree XML API

File Formats
csv: CSV File Reading and Writing
ConfigParser: Configuration file parser
robotparser: Parser for robots.txt
netrc: netrc file processing
xdrlib: Encode and decode XDR data

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library (continued)

Cryptographic Services
hashlib: Secure hashes and message digests
hmac: Keyed-Hashing for Message Authentication
md5: MD5 message digest algorithm
sha: SHA-1 message digest algorithm

File and Directory Access

os.path: Common pathname manipulations
fileinput: Iterate over lines from multiple input streams
stat: Interpreting stat() results
statvfs: Constants used with os.statvfs()
filecmp: File and Directory Comparisons
tempfile: Generate temporary files and directories
glob: UNIX style pathname pattern expansion
fnmatch: UNIX filename pattern matching
linecache: Random access to text lines
shutil: High-level file operations
dircache: Cached directory listings

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library (continued)

Data Compression and Archiving
zlib: Compression compatible with gzip
gzip: Support for gzip files
bz2: Compression compatible with bzip2
zipfile: Work with ZIP archives
tarfile: Read and write tar archive files
Data Persistence
pickle: Python object serialization
cPickle: A faster pickle
copy_reg: Register pickle support functions
shelve: Python object persistence
marshal: Internal Python object serialization
anydbm: Generic access to DBM-style databases
whichdb: Guess which DBM module created a database
dbm: Simple "cdatabase"d interface
gdbm: GNU’s reinterpretation of dbm
dbhash: DBM-style interface to the BSD database library
bsddb: Interface to Berkeley DB library
dumbdbm: Portable DBM implementation
sqlite3: DB-API 2.0 interface for SQLite databases

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library (continued)
Generic Operating System Services
os: operating system interfaces
time: Time access and conversions
optparse: More powerful command line option parser
getopt: Parser for command line options
logging: Logging facility for Python
getpass: Portable password input
curses: Terminal handling for character-cell displays
curses.ascii: Utilities for ASCII characters
curses.panel: A panel stack extension for curses.
platform: Access to underlying platform’s identifying data.
errno: Standard errno system symbols
ctypes: A foreign function library for Python.
Optional Operating System Services
select: Waiting for I/O completion
thread: Multiple threads of control
threading: Higher-level threading interface
dummy_thread: Drop-in replacement for the thread module
dummy_threading: Drop-in replacement for the threading module
mmap: Memory-mapped file support
readline: GNU readline interface
rlcompleter: Completion function for GNU readline

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library (continued)

Unix Specific Services
posix: The most common POSIX system calls
pwd: The password database
spwd: The shadow password database
grp: The group database
crypt: Function to check UNIX passwords
dl: C functions in shared objects
termios: POSIX style tty control
tty: Terminal control functions
pty: Pseudo-terminal utilities
fcntl: The fcntl() and ioctl() system calls
pipes: Interface to shell pipelines
posixfile: File-like objects with locking support
resource: Resource usage information
nis: Interface to Sun’s NIS (Yellow Pages)
syslog: UNIX syslog library routines
commands: Utilities for running commands

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library (continued)

Interprocess Communication and Networking
subprocess: Subprocess management
socket: Low-level networking interface
signal: Set handlers for asynchronous events
popen2: Subprocesses with accessible I/O streams
asyncore: Asynchronous socket handler
asynchat: Asynchronous socket command/response handler

Internet Protocols and Support

webbrowser: Convenient Web-browser controller
cgi: Common Gateway Interface support
cgitb: Traceback manager for CGI scripts
wsgiref: WSGI Utilities and Reference Implementation
urllib: Open arbitrary resources by URL
urllib2: extensible library for opening URLs
httplib: HTTP protocol client
ftplib: FTP protocol client
gopherlib: Gopher protocol client
poplib: POP3 protocol client
imaplib: IMAP4 protocol client

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library (continued)

nntplib: NNTP protocol client
smtplib: SMTP protocol client
smtpd: SMTP Server
telnetlib: Telnet client
uuid: UUID objects according to RFC 4122
urlparse: Parse URLs into components
SocketServer: A framework for network servers
BaseHTTPServer: Basic HTTP server
SimpleHTTPServer: Simple HTTP request handler
CGIHTTPServer: CGI-capable HTTP request handler
cookielib: Cookie handling for HTTP clients
Cookie: HTTP state management
xmlrpclib: XML-RPC client access
SimpleXMLRPCServer: Basic XML-RPC server
DocXMLRPCServer: Self-documenting XML-RPC server

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library (continued)
Internationalization
gettext: Multilingual internationalization services
locale: Internationalization services

Multimedia Services
audioop: Manipulate raw audio data
imageop: Manipulate raw image data
aifc: Read and write AIFF and AIFC files
sunau: Read and write Sun AU files
wave: Read and write WAV files
chunk: Read IFF chunked data
colorsys: Conversions between color systems
rgbimg: Read and write "cSGI RGB"d files
imghdr: Determine the type of an image
sndhdr: Determine type of sound file
ossaudiodev: Access to OSS-compatible audio devices

Graphical User Interfaces with Tk

Tkinter: Python interface to Tcl/Tk
Tix: Extension widgets for Tk
ScrolledText: Scrolled Text Widget
turtle: Turtle graphics for Tk

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library (continued)
Program Frameworks
cmd: Support for line-oriented command interpreters
shlex: Simple lexical analysis

Development Tools
pydoc: Documentation generator and online help system
doctest: Test interactive Python examples
unittest: Unit testing framework

The Python Profilers/Debuggers

pdb: Python Debugger
hotshot: High performance logging profiler
timeit: Measure execution time of small code snippets
trace: Trace or track Python statement execution

Python Runtime Services

sys: System-specific parameters and functions
warnings: Warning control
atexit: Exit handlers
traceback: Print or retrieve a stack traceback
__future__: Future statement definitions
gc: Collector interface
inspect: Inspect live objects
fpectl: Floating point exception control
B. Blais Numerical Computing in Python
 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Standard Library (continued)

Python Language Services
parser: Access Python parse trees
symbol: Constants used with Python parse trees
token: Constants used with Python parse trees
keyword: Testing for Python keywords
tokenize: Tokenizer for Python source
tabnanny: Detection of ambiguous indentation
pyclbr: Python class browser support
py_compile: Compile Python source files
compileall: Byte-compile Python libraries
dis: Disassembler for Python byte code
pickletools: Tools for pickle developers.
distutils: Building and installing Python modules

MS Windows Specific Services

msilib: Read and write Microsoft Installer files
msvcrt: Useful routines from the MS VC++ runtime
_winreg: Windows registry access
winsound: Sound-playing interface for Windows

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Other Libraries for Multiple Purposes

Other Libraries - cheeseshop.python.org
BeautifulSoup HTML/XML parser for quick-turnaround applications
CherryPy 3.0.1 Object-Oriented HTTP framework
Wax 0.3.19 A user-friendly layer on top of wxPython.
xlrd 0.6.1a4 Library to extract data from Microsoft Excel(tm) spreadsheet files
buzhug 0.7 A fast pure-Python database engine
psyco 1.2 Psyco, the Python specializing compiler
TurboGears 1.0.2 front-to-back rapid web development
SQLAlchemy 0.3.7 Database Abstraction Library
Shed Skin 0.0.21 An Optimizing Python-to-C++ Compiler
Golly An open source, cross-platform Game of Life simulator
pyXLWriter 0.4a2 A library for generating Excel Spreadsheets
VPython A Python module that offers real-time 3D output,

(Currently 2355 packages as of 5/08/2007)

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
In Favor of Matlab
Advantages
In Favor of Python
Extensions with Pyrex
Communication

Notable Differences in Favor of Python

Other Libraries for Multiple Purposes

Other Libraries - cheeseshop.python.org
BeautifulSoup HTML/XML parser for quick-turnaround applications
CherryPy 3.0.1 Object-Oriented HTTP framework
Wax 0.3.19 A user-friendly layer on top of wxPython.
xlrd 0.6.1a4 Library to extract data from Microsoft Excel(tm) spreadsheet files
buzhug 0.7 A fast pure-Python database engine
psyco 1.2 Psyco, the Python specializing compiler
TurboGears 1.0.2 front-to-back rapid web development
SQLAlchemy 0.3.7 Database Abstraction Library
Shed Skin 0.0.21 An Optimizing Python-to-C++ Compiler
Golly An open source, cross-platform Game of Life simulator
pyXLWriter 0.4a2 A library for generating Excel Spreadsheets
VPython A Python module that offers real-time 3D output,

(Currently 2355 packages as of 5/08/2007)

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Factorial
Advantages
Using Numpy Arrays
Extensions with Pyrex
Communication

Factorial

Extensions with Pyrex

Pyrex
Python
def factorial(int n):
def factorial(n): cdef int i
cdef double x
x=1
for i in range(1,n+1): x=1
x=x*i for i from 1 <= i <= n:
x=x*i
return x
return x

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Factorial
Advantages
Using Numpy Arrays
Extensions with Pyrex
Communication

Factorial

Extensions with Pyrex

Pyrex
Python
def factorial(int n):
def factorial(n): cdef int i
cdef double x
x=1
for i in range(1,n+1): x=1
x=x*i for i from 1 <= i <= n:
x=x*i
return x
return x

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Factorial
Advantages
Using Numpy Arrays
Extensions with Pyrex
Communication

Compiling

Setup
from pyrex_compile import *
compile(’factorial_pyrex.pyx’)

Test
In [1]:import factorial_python as python
In [2]:import factorial_pyrex as pyrex

In [3]:%timeit python.factorial(200)
1000 loops, best of 3: 319 microsec per loop

In [4]:%timeit pyrex.factorial(200)
100000 loops, best of 3: 4.17 microsec per loop

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Factorial
Advantages
Using Numpy Arrays
Extensions with Pyrex
Communication

Compiling

Setup
from pyrex_compile import *
compile(’factorial_pyrex.pyx’)

Test
In [1]:import factorial_python as python
In [2]:import factorial_pyrex as pyrex

In [3]:%timeit python.factorial(200)
1000 loops, best of 3: 319 microsec per loop

In [4]:%timeit pyrex.factorial(200)
100000 loops, best of 3: 4.17 microsec per loop

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Factorial
Advantages
Using Numpy Arrays
Extensions with Pyrex
Communication

C-API

Headers
/* Generated by Pyrex 0.9.5.1a on Sat Apr 28 11:15:48 2007 */
#include "Python.h"
#include "structmember.h"
#ifndef PY_LONG_LONG
#define PY_LONG_LONG LONG_LONG
#endif
#ifdef __cplusplus
#define __PYX_EXTERN_C extern "C"
#else
#define __PYX_EXTERN_C extern
#endif
__PYX_EXTERN_C double pow(double, double);

typedef struct {PyObject **p; char *s;} __Pyx_InternTabEntry; /*proto*/

typedef struct {PyObject **p; char *s; long n;} __Pyx_StringTabEntry; /*proto*/

static PyObject *__pyx_m;

static PyObject *__pyx_b;
static int __pyx_lineno;
static char *__pyx_filename;
static char **__pyx_f;

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Factorial
Advantages
Using Numpy Arrays
Extensions with Pyrex
Communication

C-API
Translation
int __pyx_v_n;
int __pyx_v_i;
double __pyx_v_x;
PyObject *__pyx_r;
PyObject *__pyx_1 = 0;
static char *__pyx_argnames[] = {"n",0};
if (!PyArg_ParseTupleAndKeywords(__pyx_args, __pyx_kwds, "i",
__pyx_argnames, &__pyx_v_n)) return 0;
/* "/home/bblais/tex/bryant/facdev/spring2007/src/factorial_pyrex.pyx":5 */
__pyx_v_x = 1;

/* "/home/bblais/tex/bryant/facdev/spring2007/src/factorial_pyrex.pyx":6 */
for (__pyx_v_i = 1; __pyx_v_i <= __pyx_v_n; ++__pyx_v_i) {

/* "/home/bblais/tex/bryant/facdev/spring2007/src/factorial_pyrex.pyx":7 */
__pyx_v_x = (__pyx_v_x * __pyx_v_i);
}
/* "/home/bblais/tex/bryant/facdev/spring2007/src/factorial_pyrex.pyx":9 */
__pyx_1 = PyFloat_FromDouble(__pyx_v_x); if (!__pyx_1) {__pyx_filename =
__pyx_f[0]; __pyx_lineno = 9; goto __pyx_L1;}
__pyx_r = __pyx_1;
__pyx_1 = 0;

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Factorial
Advantages
Using Numpy Arrays
Extensions with Pyrex
Communication

Watch Out!

Test
In [1]:import factorial_python as python

In [2]:import factorial_pyrex as pyrex

In [3]:%timeit python.factorial(200)
1000 loops, best of 3: 319 microsec per loop

In [4]:%timeit pyrex.factorial(200)
100000 loops, best of 3: 4.17 microsec per loop

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Factorial
Advantages
Using Numpy Arrays
Extensions with Pyrex
Communication

Watch Out!

Test
In [1]:import factorial_python as python

In [2]:import factorial_pyrex as pyrex

In [3]:%timeit python.factorial(200)
1000 loops, best of 3: 319 microsec per loop

In [4]:%timeit pyrex.factorial(200)
100000 loops, best of 3: 4.17 microsec per loop

In [5]:pyrex.factorial(200)
Out[5]:inf

In [6]:python.factorial(200)
Out[6]:7886578673647905035523632139321850622951359776871732632947425332443
59449963403342920304284011984623904177212138919638830257642790242637105061
92662495282993111346285727076331723739698894392244562145166424025403329186
41312274282948532775242424075739032403212574055795686602260319041703240623
51700858796178922222789623703897374720000000000000000000000000000000000000
000000000000L

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Factorial
Advantages
Using Numpy Arrays
Extensions with Pyrex
Communication

Function using Array

Pyrex Code
cimport c_python
cimport c_numpy
c_numpy.import_array() # Numpy must be initialized

def spam(c_numpy.ndarray A):

cdef double *p
cdef double result
cdef int i,N,nd

nd=A.nd # number of dimensions

N=1
for i from 0<=i<nd: # calculate number of elements
N=N*A.dimensions[i]

p=<double *>A.data

result=0.0
for i from 0<=i<N:
result=result+p[i]**2

return result

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Factorial
Advantages
Using Numpy Arrays
Extensions with Pyrex
Communication

Testing
In [1]:from numpy.random import rand
In [2]:from pyrex_numpy import spam
In [3]:a=rand(2,3,4)
In [4]:a
Out[4]:
array([[[ 0.41275586, 0.40248059, 0.52365634, 0.13457172],
[ 0.10361721, 0.07592018, 0.50031702, 0.65126816],
[ 0.09734859, 0.82231387, 0.74795067, 0.48530395]],

[[ 0.17096585, 0.42510408, 0.4848095 , 0.12744915],

[ 0.49256875, 0.1358942 , 0.12986233, 0.86068033],
[ 0.10222339, 0.46645587, 0.82551456, 0.54402251]]])

In [5]:a.shape
Out[5]:(2, 3, 4)
In [6]:spam(a)
Out[6]:5.481696128900011
In [7]:(a**2).sum()
Out[7]:5.4816961289

In [8]:%timeit spam(a)
1000000 loops, best of 3: 465 ns per loop

In [9]:%timeit (a**2).sum()
10000 loops, best of 3: 31.2 microsec per loop

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Matlab and R
Advantages
Excel
Extensions with Pyrex
Communication

mlabwrap.py

Communicating with
Matlab
from mlabwrap import mlab
from numpy import *

xx = arange(-2*pi, 2*pi, 0.2)

X,Y=mlab.meshgrid(xx,xx,nout=2)
mlab.surf(sin(X)*cos(Y))

mlab.xlabel(’This’)
mlab.ylabel(’That’)
mlab.title(’Something Interesting’)
mlab.zlabel(’Value’)

a=mlab.svd(array([[1,2], [1,3]]))

print a

# Run:
In [44]:run test_mlab.py
[[ 3.86432845]
[ 0.25877718]]

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Matlab and R
Advantages
Excel
Extensions with Pyrex
Communication

rpy.py
Communicating with R
# Simple script for drawing the chi-squared density
#
from rpy import *

degrees = 4
grid = r.seq(0, 10, length=100)
values = [r.dchisq(x, degrees) for x in grid]
r.par(ann=0)
r.plot(grid, values, type=’lines’)

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Matlab and R
Advantages
Excel
Extensions with Pyrex
Communication

xlrd.py

Reading Excel Spreadsheets

import xlrd
book = xlrd.open_workbook("master.xls")
print "The number of worksheets is", book.nsheets
print "Worksheet name(s):", book.sheet_names()
sh = book.sheet_by_index(0)
print sh.name, sh.nrows, sh.ncols
print "Cell D30 is", sh.cell_value(rowx=29, colx=3)
for rx in range(sh.nrows):
print sh.row(rx)

B. Blais Numerical Computing in Python

 Introduction
Comparison With Matlab
Matlab and R
Advantages
Excel
Extensions with Pyrex
Communication

Conclusions

Questions?

B. Blais Numerical Computing in Python