The Python Programming Language¶
Invented in 1991 by this guy: 
Python Vital Statistics¶
- Python is multi-paradigm
- You can write Python code that is object-oriented, procedural, or functional (or any combination)
- Python is dynamically typed & interpreted
- Type and syntax are checked at runtime, there is no compile/link stage
- Python is named after the Flying Circus, not the snake
- Guido van Rossum's official title is Benevolent Dictator For Life
In [1]:
import this
The Scientific Python Stack¶
Numpy for Array Operations¶
Scipy for scientific algorithms¶
Matplotlib for Plotting¶
Jupyter/IPython Notebook for Interactivity¶
Python Syntax¶
In [2]:
#Variable type is inferred in python ("Duck Typing")
my_int = 3
my_string = "foo"
print "Type of my_int:", type(my_int)
print "Type of my_string:", type(my_string)
In [3]:
#Python operators can mean different things for different types
print my_int*3
print my_string*3
In [4]:
#In python, we don't need {}, ;, and indentation matters
if my_int > 0:
print my_int, "is positive"
else:
print "Uh-oh"
print "And now we are back!"
In [5]:
#Python reads like english, so boolean operations are just words
print True and False
print True or False
print not True
In [6]:
#Functions in python are familiar, but because python is duck-typed,
#we don't have return type
def is_even(number):
return (number % 2) == 0
print 'is 3 even?', is_even(3)
print 'is 100 even?', is_even(100)
In [7]:
#Even though type is inferred, python is strongly-typed: certain
#operations are forbidden
print "1"+1
In [8]:
#Python is object-oriented: Every variable is an object,
#and objects can have methods
alist = [1,2,3]
print alist
alist.reverse()
print alist
In [9]:
#This is how the operator overloading works
print dir(alist)
Syntactic Sugar¶
Nice arrays/lists¶
In [10]:
#Define an array
a = [1,2,3,4]
In [11]:
#Print the first & last element
print a[0], a[-1]
In [12]:
#Add some more elements to the list
a = a + [5,6,7]
print a
In [13]:
#Get a subset of the list
print a[2:5]
In [14]:
#Find the length of a list
print len(a)
In [15]:
#Print every second element
print a[0:-1:2]
In [16]:
#2D Arrays are easy too
b = [a,a,a]
print b
print b[1]
print b[2][3]
Dictionaries¶
In [17]:
#Dicts are a kind of datastructure called a Hash Table
codon = {'UUU':'Phenylalanine', 'CUU':'Isoleucine', 'GGU':'Glycine', 'AGU':'Serine'}
#Lookup in a dict runs in constant time
print 'UUU encodes ', codon['UUU']
#Dicts have keys & Values
print 'The Keys are:', codon.keys()
print 'The Values are:', codon.values()
Comprehensions¶
In [18]:
#Rather than build a list with a loop, you can use a list-comprehension
squares = []
for i in range(10):
squares.append(i*i)
print "With a loop", squares
squares = [i*i for i in range(10)]
print "With a comprehension", squares
In [19]:
#Dicts also have comprehensions
cubes = {i:i*i*i for i in range(10)}
print cubes.keys()
print cubes.values()
print "3 cubed is", cubes[3]
Named & default arguments¶
In [20]:
#Python function arguments each have a name, but they
#can also be assigned a default value
def get_last(inlist, num=1):
return inlist[-num:]
print get_last([1,2,3,4])
print get_last([1,2,3,4], 2)
print get_last(inlist=[1,2,3,4])
print get_last(inlist=[1,2,3,4], num=3)
Plotting in Python is Dead Simple¶
Matplotlib is the de facto standard¶
In [22]:
from matplotlib.pyplot import *
from numpy import *
In [23]:
xvals = arange(0,2*pi,1e-3)
yvals = sin(xvals)
plot(xvals,yvals)
xlabel('x')
ylabel('sin(x)')
Out[23]:
In [24]:
#Let's test the central limit theorem
randnums = [mean(random_integers(0,10,1000)) for i in range(10000)]
null = hist(randnums,bins=50)
In [25]:
#Even 3d plots work nicely
from mpl_toolkits.mplot3d import axes3d
ax = gcf().add_subplot(111, projection='3d')
X, Y, Z = axes3d.get_test_data(0.05)
ax.plot_wireframe(X,Y,Z, rstride=10, cstride=10)
Out[25]:
