Python Language : Simplicity and Expressivity¶

Kastoria TEIWM April 2019¶

In [2]:

PythonLogo

Out[2]:

Main Python strengths¶

Simplicity
Concision (~1/3 of equivalent program in Java or C)
Portability
Batteries Included

Other caracteristics¶

python scripts
Interpreted
Object Oriented
Weak types (vs Java)
Open Source

Python users¶

Google
YouTube
Dropbox
Internet Boxes
Scientific tool
Web mining
Natural language processing
Deep Learning (TensorFlow, PyTorch, Scikit-Learn)

Inconvenients¶

Rather slow
Hard roadmap from Python2 to Python3

Environnements¶

Commandline REPL (Read Eval Print Loop)
Dedicated editor(IDLE, Emacs, PyCharm, SublimeText, Atom, VSCode, vim, etc.)
Jupyter notebooks

Jupyter notebook:¶

Browser is the IDE
Easy debug
Comments / Illustrations included

Practise¶

Launch the notebook
Choose the Kernel
Start coding in a cell
Press >| (RUN)
or Shift ENTER

Various Kernels available¶

Python
virtualenvs
JavaScript
SQL
Octave

Notebook in a notebook¶

In [3]:

from IPython.display import HTML
HTML('<iframe src=http://nbviewer.jupyter.org/github/ipython/ipython/blob/master/examples/IPython%20Kernel/SymPy.ipynb width=700 height=350></iframe>')

/opt/local/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/IPython/core/display.py:689: UserWarning: Consider using IPython.display.IFrame instead
  warnings.warn("Consider using IPython.display.IFrame instead")

Out[3]:

Let's start coding in Python¶

Basic types¶

Integers
Floats
Booleans
String
None (like void in C++)

Elementary data structures¶

Lists
Tuples
Dictionnaries

numerical examples¶

In [4]:

1 + 3

Out[4]:

In [5]:

5 * 7

Out[5]:

In [6]:

5 / 2

Out[6]:

2.5

In [7]:

5 // 2

Out[7]:

In [8]:

0.1 + 0.1 + 0.1 - 0.3
# What's going on ?

Out[8]:

5.551115123125783e-17

Booleans¶

In [9]:

True and True

Out[9]:

True

In [10]:

True and False

Out[10]:

False

In [11]:

True or False

Out[11]:

True

In [12]:

( 2 > 3 ) and ( 1 < 2)

Out[12]:

False

In [13]:

( 1 < 2 < 5 )

Out[13]:

True

Strings¶

Declare and print¶

In [14]:

mystring = 'qwerty'
print(mystring)

qwerty

In [15]:

otherstring = "a nice string"
print(otherstring)

a nice string

Length¶

In [16]:

empty = ''
len(empty)

Out[16]:

In [17]:

len(otherstring)

Out[17]:

In [18]:

len(None)

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-18-96b07938784c> in <module>
----> 1 len(None)

TypeError: object of type 'NoneType' has no len()

Operations on strings¶

In [19]:

S1 = "Γειάσου "
S2 = "Κόσμε !"
print(S1+S2)

Γειάσου Κόσμε !

We can't do anything:

In [20]:

S1*S2

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-20-e4eb3fd43725> in <module>
----> 1 S1*S2

TypeError: can't multiply sequence by non-int of type 'str'

But still

In [21]:

S1 * 3

Out[21]:

'Γειάσου Γειάσου Γειάσου '

In [22]:

# or
print('*'*60)

************************************************************

Extract substrings¶

In [23]:

S = 'azertyuiop'
print(S[1])
print(S[1:5])

z
zert

and what about this ?

In [24]:

print(S[-1])

print(S[-2])

S[::-1]

p
o

Out[24]:

'poiuytreza'

In [28]:

S = "Hello, the weather is rainy today, isn't it ? "
S.split(',')

Out[28]:

['Hello', ' the weather is rainy today', " isn't it ? "]

search and replace in strings¶

In [29]:

S.find('rainy')
# Returns the index of 'rainy' in the string S

Out[29]:

In [30]:

S.replace('rainy','nice')

Out[30]:

"Hello, the weather is nice today, isn't it ? "

In [31]:

S = "ahoouutch  "
S.rstrip()
# cleaning spaces

Out[31]:

'ahoouutch'

Import interesting libs¶

In [32]:

# Let's import networkx to deal with graphs
import networkx as nx
# Let's create a digraph G
G = nx.DiGraph()
# and add 3 nodes to G
G.add_nodes_from('ABC')
# add an edge from A to B
G.add_edge('A','B')
# and an other from A to C
G.add_edge('A','C')
print(G)
# We can give some pos
pos = { 'A' : (0,0), 'B' : (1,1), 'C' : (-1,1) }
# and draw the graph
nx.draw(G, pos, edge_color='b',node_color='y')

Manipulate data in csv files ¶

In [33]:

import pandas

In [ ]:

%%bash
mkdir data

In [35]:

%%file data/data.csv
Langage,Strengths, Weaknesses
Java, Structured, Verbose
JavaScript,'Ubiquitous','Weard'
PHP, Simple,'Not very concise'
Python,Powerful,'2to3'

Overwriting data/data.csv

In [36]:

# Read again file
df = pandas.read_csv('data/data.csv')

In [37]:

df

Out[37]:

	Langage	Strengths	Weaknesses
0	Java	Structured	Verbose
1	JavaScript	'Ubiquitous'	'Weard'
2	PHP	Simple	'Not very concise'
3	Python	Powerful	'2to3'

.. et manipulate texts¶

In [38]:

import nltk

In [39]:

phrase="today, the weather is mild and I would like to go around the lake"
tokens=nltk.word_tokenize(phrase)
tokens
# Try completions nltk. ...

Out[39]:

['today',
 ',',
 'the',
 'weather',
 'is',
 'mild',
 'and',
 'I',
 'would',
 'like',
 'to',
 'go',
 'around',
 'the',
 'lake']

We can use bigger texts

In [40]:

from urllib import request
url='http://www.gutenberg.org/cache/epub/1497/pg1497.txt'
response = request.urlopen(url)
republic = response.read().decode('utf8')
type(republic)

# to go through a proxy:
# proxies = {'http': 'http://www.myproxy.com:3128'}
# request.ProxyHandler(proxies)

Out[40]:

str

In [41]:

len(republic)

Out[41]:

In [42]:

republic[231900:232071]

Out[42]:

'And the steps of the ladder leading up to this highest or\r\nuniversal existence are the mathematical sciences, which also contain\r\nin themselves an element of the universal'

In [ ]: