LOGO

UNIV/POLTEK

Python Basics
Function, Data Structures, Data Manipulation

[Link]
LOGO
UNIV/POLTEK

Functional Programming tools

• Python allows for the use of a special kind of function, a lambda function.
• Lambda functions are small, anonymous functions based on the lambda abstractions that
appear in many functional languages.
• Python can support many different programming paradigms including functional
programming.
Right now, we’ll take a look at some of the handy functional tools provided by Python.

[Link]
LOGO
UNIV/POLTEK

Lambda functions

• Lambda functions within Python.

• Use the keyword lambda instead of def. >>> def f(x):
• Can be used wherever function objects are used. ... return x**2
• Restricted to one expression. ...
• Typically used with functional programming tools. >>> f(8)
64
>>> g = lambda x: x**2
>>> g(8)
64

[Link]
LOGO
UNIV/POLTEK

Functional programming tools

def even(x):
• Filter if x % 2 == 0:
• • filter(function, iterable). Filters method return True
constructs an iterator from elements of an else:
iterable for which a function returns true. return False
iterable which is to be filtered, could be
sets, lists, tuples, or containers of any list(filter(even, range(0,30)))
iterators
• • Returns an iterator that passed the [0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28]
function check for each element in the
iterable.

[Link]
LOGO
UNIV/POLTEK

Functional programming tools

• Map
def square(x):
• • map(function, sequence) applies
function to each item in sequence and return x**2
returns the results as a list.
list(map(square, range(0,11)))
• • Multiple arguments can be provided if
the function supports it.
[0, 1, 4, 9, 16, 25, 36, 49, 64, 81, 100]

[Link]
LOGO
UNIV/POLTEK

Functional programming tools

• Map def expo(x, y):

• • map(function, sequence) return x**y
applies function to each item in
sequence and returns the results list(map(expo, range(0,5), range(0,5)))
as a list.
• • Multiple arguments can be
provided if the function supports [1, 1, 4, 27, 256]
it.

[Link]
LOGO
UNIV/POLTEK

Functional programming tools

• Reduce
• • reduce(function, sequence) returns a import functools
single value computed as the result of
performing function on the first two def fact(x, y):
items, then on the result with the next return x*y
item, etc.
• • There’s an optional third argument print([Link](fact,
which is the starting value. range(1,5)))

24

[Link]
LOGO
UNIV/POLTEK

Functional programming tools

• We can combine lambda abstractions with functional programming tools. This is especially
useful when our function is small – we can avoid the overhead of creating a function
definition for it by essentially defining it in-line.

>>> list(map(lambda x: x**2, range(0,11)))

[0, 1, 4, 9, 16, 25, 36, 49, 64, 81, 100]

[Link]
LOGO
UNIV/POLTEK

More Data Structures

• Lists
• Slicing
• Stacks and Queues
• Tuples
• Sets and Frozensets
• Dictionaries
• How to choose a data structure.
• Collections
• Deques and OrderedDicts

[Link]
LOGO
UNIV/POLTEK

When to use Lists

• When you need a non-homogeneous collection of elements.
• When you need the ability to order your elements.
• When you need the ability to modify or add to the collection.
• When you don't require elements to be indexed by a custom value.
• When you need a stack or a queue.
• When your elements are not necessarily unique.

[Link]
LOGO
UNIV/POLTEK

Creating lists
• To create a list in Python, we can use bracket notation to either create an empty list or an
initialized list.
mylist1 = [] # Creates an empty list
mylist2 = [expression1, expression2, ...]
mylist3 = [expression for variable in sequence]

Example:
mylist2 = [0,1,4,9,16]
mylist3 = [i for i*i in range(5)]
• The first two are referred to as list displays, where the last example is a list comprehension.

[Link]
LOGO
UNIV/POLTEK

Creating lists
• We can also use the built-in list constructor to create a new list.

mylist1 = list()
mylist2 = list(sequence)
mylist3 = list(expression for variable in sequence)
• The sequence argument in the second example can be any kind of sequence object or
iterable. If another list is passed in, this will create a copy of the argument list.

[Link]
LOGO
UNIV/POLTEK

Creating lists
• Note that you cannot create a new different list without initialization.

# mylist1 and mylist2 point to the same list

mylist1 = mylist2 = []

# mylist3 and mylist4 point to the same list

mylist3 = []
mylist4 = mylist3

mylist5 = []; mylist6 = [] # different lists

[Link]
LOGO
UNIV/POLTEK

Accessing list elements

• If the index of the desired element is known, you can simply use bracket notation to index
into the list.
>>> mylist = [34,67,45,29]
>>> mylist[2] 45
45

• If the index is not known, use the index() method to find the first index of an item. An
exception will be raised if the item cannot be found.
>>> mylist = [34,67,45,29]
>>> [Link](67) 1
1

[Link]
LOGO
UNIV/POLTEK

Slicing and sliding

• The length of the list is accessible through len(mylist).
• Slicing is an extended version of the indexing operator and can be used to grab sublists.

mylist[start:end] # items start to end-1

mylist[start:] # items start to end of the array
mylist[:end] # items from beginning to end-1
mylist[:] # a copy of the whole array

• You may also provide a step argument with any of the slicing constructions above.

mylist[start:end:step] # start to end-1, by step

[Link]
LOGO
UNIV/POLTEK

Slicing and sliding

• The start or end arguments may be negative numbers, indicating a count from the end of the
array rather than the beginning. This applies to the indexing operator.

mylist[-1] # last item in the array

mylist[-2:] # last two items in the array
mylist[:-2] # everything except the last two items

• Some examples:
mylist = [34, 56, 29, 73, 19, 62]
mylist[-2] # yields 19
mylist[-4::2] # yields [29, 19]

[Link]
LOGO
UNIV/POLTEK

Inserting/removing elements
• To add an element to an existing list, use the append() method.

>>> mylist = [34, 56, 29, 73, 19, 62]

>>> [Link](47)
>>> mylist
[34, 56, 29, 73, 19, 62, 47]

• Use the extend() method to add all of the items from another list.
>>> mylist = [34, 56, 29, 73, 19, 62]
>>> [Link]([47,81])
>>> mylist
[34, 56, 29, 73, 19, 62, 47, 81]
[Link]
LOGO
UNIV/POLTEK

Inserting/removing elements
• Use the insert(pos, item) method to insert an item at the given position. You may also use
negative indexing to indicate the position.
>>> mylist = [34, 56, 29, 73, 19, 62]
>>> [Link](2,47)
>>> mylist
[34, 56, 47, 29, 73, 19, 62]
• Use the remove() method to remove the first occurrence of a given item. An exception will be
raised if there is no matching item in the list.
>>> mylist = [34, 56, 29, 73, 19, 62]
>>> [Link](29)
>>> mylist
[34, 56, 73, 19, 62]
[Link]
LOGO
UNIV/POLTEK

Lists as stacks
• You can use lists as a quick stack data structure.
• The append() and pop() methods implement a Last In, First Out (LIFO) structure.
• The pop(index) method will remove and return the item at the specified index. If no
index is specified, the last item is popped from the list.
>>> stack = [34, 56, 29, 73, 19, 62]
>>> [Link](47)
>>> stack
[34, 56, 29, 73, 19, 62, 47]
>>> [Link]()
47
>>> stack
[34, 56, 29, 73, 19, 62]
[Link]
LOGO
UNIV/POLTEK

Lists as queues

• Lists can be used as queues natively >>> from collections import deque
since insert() and pop() both support >>> queue = deque([35, 19, 67])
indexing. However, while appending >>> [Link](42)
and popping from a list are fast,
inserting and popping from the >>> [Link](23)
beginning of the list are slow >>> [Link]()
(especially with large lists. Why is 35
this?). >>> [Link]()
• Use the special deque object from 19
the collections module.
>>> queue
deque([67, 42, 23])

[Link]
LOGO
UNIV/POLTEK

Other operations
• The count(x) method will give you the number of occurrences of item x within the list.

>>> mylist = ['a', 'b', 'c', 'd', 'a', 'f', 'c']

>>> [Link]('a')
2

• The sort() and reverse() methods sort and reverse >>> mylist = [5, 2, 3, 4, 1]
the list in place. The sorted(mylist) and
reversed(mylist) built-in functions will return a >>> [Link]()
sorted and reversed copy of the list, respectively. >>> mylist
[1, 2, 3, 4, 5]
>>> [Link]()
>>> mylist
[5, 4, 3, 2, 1]
[Link]
LOGO
UNIV/POLTEK

Custom sorting
we can sort the list as follow (default = ascending)
>>> mylist = ['b', 'A', 'D', 'c']
>>> [Link](key = [Link])
>>> mylist
['A', 'b', 'c', 'D']

For descending
>>> mylist = ['b', 'A', 'D', 'c']
>>> [Link](key = [Link], reverse = True)
>>> mylist
[‘D', ‘c', ‘b', ‘A']
[Link]() is a built-in string method.
[Link]
LOGO
UNIV/POLTEK

When to use sets

• When the elements must be unique.
• When you need to be able to modify or add to the collection.
• When you need support for mathematical set operations.
• When you don't need to store nested lists, sets, or dictionaries as elements.

[Link]
LOGO
UNIV/POLTEK

Creating sets
• Create an empty set with the set constructor.
myset = set()
myset2 = set([]) # both are empty sets

• Create an initialized set with the set constructor or the { } notation. Do not use empty curly
braces to create an empty set – you’ll get an empty dictionary instead.

myset = set(sequence)
myset2 = {expression for variable in sequence}

[Link]
LOGO
UNIV/POLTEK

Hashable items
• The way a set detects non-unique elements is by indexing the data in memory, creating a
hash for each element. This means that all elements in a set must be hashable.
• All of Python’s immutable built-in objects are hashable, while no mutable containers (such as
lists or dictionaries) are. Objects which are instances of user-defined classes are also hashable
by default.

[Link]
LOGO
UNIV/POLTEK

Mutable operations
>>> myset = {x for x in 'abracadabra'}
• The following operations are not >>> myset
available for frozensets. set(['a', 'b', 'r', 'c', 'd'])
• The add(x) method will add element x >>> [Link]('y')
to the set if it’s not already there. The >>> myset
remove(x) and discard(x) methods will set(['a', 'b', 'r', 'c', 'd', 'y'])
remove x from the set. >>> [Link]('a')
• The pop() method will remove and >>> myset
return an arbitrary element from the set(['b', 'r', 'c', 'd', 'y'])
set. Raises an error if the set is empty. >>> [Link]()
• The clear() method removes all 'b'
elements from the set. >>> myset
set(['r', 'c', 'd', 'y'])

[Link]
LOGO
UNIV/POLTEK

Mutable operations continued

set |= other | ...

Update the set, adding elements from all others.

set &= other & ...

Update the set, keeping only elements found in it and all others.

set -= other | ...

Update the set, removing elements found in others.

set ^= other
Update the set, keeping only elements found in either set, but not in both.

[Link]
LOGO
UNIV/POLTEK

Mutable operations continued

>>> s1 = set('abracadabra')
>>> s2 = set('alacazam')
>>> s1
set(['a', 'b', 'r', 'c', 'd'])
>>> s2
set(['a', 'l', 'c', 'z', 'm'])
>>> s1 |= s2
>>> s1
set(['a', 'b', 'r', 'c', 'd', 'l', 'z', 'm'])
>>> s1 = set('abracadabra')
>>> s1 &= s2
>>> s1
set(['a', 'c'])
[Link]
LOGO
UNIV/POLTEK

Set operations
• The following operations are available for both set and frozenset types.
• Comparison operators >=, <= test whether a set is a superset or subset, respectively, of some
other set. The > and < operators check for proper supersets/subsets.

>>> s1 = set('abracadabra')
>>> s2 = set('bard')
>>> s1 >= s2
True
>>> s1 > s2
True
>>> s1 <= s2
False
[Link]
LOGO
UNIV/POLTEK

Set operations
• Union: set | other | …
• Return a new set with elements from the set and all others.
• Intersection: set & other & …
• Return a new set with elements common to the set and all others.
• Difference: set – other – …
• Return a new set with elements in the set that are not in the others.
• Symmetric Difference: set ^ other
• Return a new set with elements in either the set or other but not both.

[Link]
LOGO
UNIV/POLTEK

Set operations
>>> s1 = set('abracadabra')
>>> s1
set(['a', 'b', 'r', 'c', 'd'])
>>> s2 = set('alacazam')
>>> s2
set(['a', 'l', 'c', 'z', 'm'])
>>> s1 | s2
set(['a', 'b', 'r', 'c', 'd', 'l', 'z', 'm'])
>>> s1 & s2
set(['a', 'c'])
>>> s1 - s2
set(['b', 'r', 'd'])
>>> s1 ^ s2
set(['b', 'r', 'd', 'l', 'z', 'm'])
[Link]
LOGO
UNIV/POLTEK

Other operations
• [Link]() returns a shallow copy of the set s.
• [Link](other) returns True if set s has no elements in common with set other.
• [Link](other) returns True if set s is a subset of set other.
• len, in, and not in are also supported.

[Link]
LOGO
UNIV/POLTEK

When to use tuples

• When storing elements that will not need to be changed.
• When performance is a concern.
• When you want to store your data in logical immutable pairs, triples, etc.

[Link]
LOGO
UNIV/POLTEK

Constructing tuples
• An empty tuple can be created with an empty set of parentheses.
• Pass a sequence type object into the tuple() constructor.
• Tuples can be initialized by listing comma-separated values. These do not need to be in
parentheses but they can be.
• One quirk: to initialize a tuple with a single value, use a trailing comma.

>>> t1 = (1, 2, 3, 4)
>>> t2 = "a", "b", "c", "d"
>>> t3 = ()
>>> t4 = ("red", )

[Link]
LOGO
UNIV/POLTEK

Tuple operations
• Tuples are very similar to lists and support a lot of the same operations.
• Accessing elements: use bracket notation (e.g. t1[2]) and slicing.
• Use len(t1) to obtain the length of a tuple.
• The universal immutable sequence type operations are all supported by tuples.
• +, *
• in, not in
• min(t), max(t), [Link](x), [Link](x)

[Link]
LOGO
UNIV/POLTEK

Packing/unpacking
• Tuple packing is used to “pack” a collection of items into a tuple. We can unpack a tuple using
Python’s multiple assignment feature.

>>> s = "Susan", 19, "CS" # tuple packing

>>> name, age, major = s # tuple unpacking
>>> name
'Susan'
>>> age
19
>>> major
'CS'

[Link]
LOGO
UNIV/POLTEK

When to use dictionaries

• When you need to create associations in the form of key:value pairs.
• When you need fast lookup for your data, based on a custom key.
• When you need to modify or add to your key:value pairs.

[Link]
LOGO
UNIV/POLTEK

Constructing a dictionary
• Create an empty dictionary with empty curly braces or the dict() constructor.
• You can initialize a dictionary by specifying each key:value pair within the curly braces.
• Note that keys must be hashable objects.

>>> d1 = {}
>>> d2 = dict() # both empty
>>> d3 = {"Name": "Susan", "Age": 19, "Major": "CS"}
>>> d4 = dict(Name="Susan", Age=19, Major="CS")
>>> d5 = dict(zip(['Name', 'Age', 'Major'], ["Susan", 19, "CS"]))
>>> d6 = dict([('Age', 19), ('Name', "Susan"), ('Major', "CS")])

Note: zip takes two equal-length collections and merges their corresponding elements into tuples.
[Link]
LOGO
UNIV/POLTEK

Accessing the dictionary

• To access a dictionary, simply index the dictionary by the key to obtain the value. An
exception will be raised if the key is not in the dictionary.

>>> d1 = {'Age':19, 'Name':"Susan", 'Major':"CS"}

>>> d1['Age']
19
>>> d1['Name']
'Susan'

[Link]
LOGO
UNIV/POLTEK

Updating a dictionary
• Simply assign a key:value pair to modify it or add a new pair. The del keyword can be used to
delete a single key:value pair or the whole dictionary. The clear() method will clear the
contents of the dictionary.

>>> d1 = {'Age':19, 'Name':"Susan", 'Major':"CS"}

>>> d1['Age'] = 21
>>> d1['Year'] = "Junior"
>>> d1
{'Age': 21, 'Name': 'Susan', 'Major': 'CS', 'Year': 'Junior'}
>>> del d1['Major']
>>> d1
{'Age': 21, 'Name': 'Susan', 'Year': 'Junior'}
>>> [Link]()
>>> d1
{}
[Link]
LOGO
UNIV/POLTEK

Built-in dictionary methods

>>> d1 = {'Age':19, 'Name':"Susan", 'Major':"CS"}
>>> d1.has_key('Age') # True if key exists
True
>>> d1.has_key('Year') # False otherwise
False
>>> [Link]() # Return a list of keys
['Age', 'Name', 'Major']
>>> [Link]() # Return a list of key:value pairs
[('Age', 19), ('Name', 'Susan'), ('Major', 'CS')]
>>> [Link]() # Returns a list of values
[19, 'Susan', 'CS']

Note: in, not in, pop(key),and popitem()are also supported.

[Link]
LOGO
UNIV/POLTEK

Ordered dictionary
Dictionaries do not remember the order in which keys were inserted. An ordered dictionary
implementation is available in the collections module. The methods of a regular dictionary are
all supported by the OrderedDict class.

An additional method supported by OrderedDict is the following:

[Link](last=True) # pops items in LIFO order

[Link]
LOGO
UNIV/POLTEK

Ordered dictionary
>>> # regular unsorted dictionary
>>> d = {'banana': 3, 'apple': 4, 'pear': 1, 'orange': 2}

>>> # dictionary sorted by key

>>> from collections import OrderedDict
>>> OrderedDict(sorted([Link](), key=lambda t: t[0]))
OrderedDict([('apple', 4), ('banana', 3), ('orange', 2), ('pear', 1)])

>>> # dictionary sorted by value

>>> OrderedDict(sorted([Link](), key=lambda t: t[1]))
OrderedDict([('pear', 1), ('orange', 2), ('banana', 3), ('apple', 4)])

>>> # dictionary sorted by length of the key string

>>> OrderedDict(sorted([Link](), key=lambda t: len(t[0])))
OrderedDict([('pear', 1), ('apple', 4), ('orange', 2), ('banana', 3)])
[Link]
LOGO
UNIV/POLTEK

Python Libraries for Data Science

Many popular Python toolboxes/libraries:

• NumPy
• SciPy
• Pandas
• SciKit-Learn

Visualization libraries
• matplotlib
• Seaborn

and many more …

44
[Link]
LOGO
UNIV/POLTEK

Python Libraries for Data Science

NumPy:
 introduces objects for multidimensional arrays and matrices, as well as functions that allow to easily
perform advanced mathematical and statistical operations on those objects

 provides vectorization of mathematical operations on arrays and matrices which significantly improves
the performance

 many other python libraries are built on NumPy

Link: [Link]

45
[Link]
LOGO
UNIV/POLTEK

Python Libraries for Data Science

SciPy:
 collection of algorithms for linear algebra, differential equations, numerical integration, optimization,
statistics and more

 part of SciPy Stack

 built on NumPy

Link: [Link]

46
[Link]
LOGO
UNIV/POLTEK

Python Libraries for Data Science

Pandas:
 adds data structures and tools designed to work with table-like data (similar to Series and Data Frames in
R)

 provides tools for data manipulation: reshaping, merging, sorting, slicing, aggregation etc.

 allows handling missing data

Link: [Link]

47
[Link]
LOGO
UNIV/POLTEK

Python Libraries for Data Science

SciKit-Learn:
 provides machine learning algorithms: classification, regression, clustering, model validation etc.

 built on NumPy, SciPy and matplotlib

Link: [Link]

48
[Link]
LOGO
UNIV/POLTEK

Python Libraries for Data Science

matplotlib:
 python 2D plotting library which produces publication quality figures in a variety of hardcopy formats

 a set of functionalities similar to those of MATLAB

 line plots, scatter plots, barcharts, histograms, pie charts etc.

 relatively low-level; some effort needed to create advanced visualization

Link: [Link]

49
[Link]
LOGO
UNIV/POLTEK

Python Libraries for Data Science

Seaborn:
 based on matplotlib

 provides high level interface for drawing attractive statistical graphics

 Similar (in style) to the popular ggplot2 library in R

Link: [Link]

50
[Link]
LOGO
UNIV/POLTEK

Loading Python Libraries

In [ ]: #Import Python Libraries
import numpy as np
import scipy as sp
import pandas as pd
import matplotlib as mpl
import seaborn as sns

Press Shift+Enter to execute the jupyter cell

51
[Link]
LOGO
UNIV/POLTEK

Reading data using pandas

In [ ]: #Read csv file
df = pd.read_csv("[Link]

Note: The above command has many optional arguments to fine-tune the data import process.

There is a number of pandas commands to read other data formats:

pd.read_excel('[Link]',sheet_name='Sheet1', index_col=None, na_values=['NA'])

pd.read_stata('[Link]')
pd.read_sas('myfile.sas7bdat')
pd.read_hdf('myfile.h5','df')

52
[Link]
LOGO
UNIV/POLTEK

Exploring data frames

In [3]: #List first 5 records
[Link]()

Out[3]:

53
[Link]
LOGO
UNIV/POLTEK

Hands-on exercises

 Try to read the first 10, 20, 50 records;

 Can you guess how to view the last few records; Hint:

54
[Link]
LOGO
UNIV/POLTEK

Data Frame data types

Pandas Type Native Python Type Description
object string The most general dtype. Will be
assigned to your column if column
has mixed types (numbers and
strings).
int64 int Numeric characters. 64 refers to
the memory allocated to hold this
character.
float64 float Numeric characters with decimals.
If a column contains numbers and
NaNs(see below), pandas will
default to float64, in case your
missing value has a decimal.
datetime64, timedelta[ns] N/A (but see the datetime module Values meant to hold time data.
in Python’s standard library) Look into these for time series
experiments.

55
[Link]
LOGO
UNIV/POLTEK

Data Frame data types

In [4]: #Check a particular column type
df['salary'].dtype

Out[4]: dtype('int64')

In [5]: #Check types for all the columns

[Link]

Out[4]: rank object

discipline object
phd int64
service int64
sex object
salary int64
dtype: object
56
[Link]
LOGO
UNIV/POLTEK

Data Frames attributes

Python objects have attributes and methods.

[Link] description
dtypes list the types of the columns

columns list the column names

axes list the row labels and column names

ndim number of dimensions

size number of elements

shape return a tuple representing the dimensionality

values numpy representation of the data

57
[Link]
LOGO
UNIV/POLTEK

Hands-on exercises

 Find how many records this data frame has;

 How many elements are there?

 What are the column names?

 What types of columns we have in this data frame?

58
[Link]
LOGO
UNIV/POLTEK

Data Frames methods

Unlike attributes, python methods have parenthesis.
All attributes and methods can be listed with a dir() function: dir(df)

[Link]() description
head( [n] ), tail( [n] ) first/last n rows

describe() generate descriptive statistics (for numeric columns only)

max(), min() return max/min values for all numeric columns

mean(), median() return mean/median values for all numeric columns

std() standard deviation

sample([n]) returns a random sample of the data frame

dropna() drop all the records with missing values

59
[Link]
LOGO
UNIV/POLTEK

Hands-on exercises

 Give the summary for the numeric columns in the dataset

 Calculate standard deviation for all numeric columns;

 What are the mean values of the first 50 records in the dataset? Hint: use

head() method to subset the first 50 records and then calculate the mean

60
[Link]
LOGO
UNIV/POLTEK

Selecting a column in a Data Frame

Method 1: Subset the data frame using column name:
df['sex']

Method 2: Use the column name as an attribute:

[Link]

Note: there is an attribute rank for pandas data frames, so to select a column with a name
"rank" we should use method 1.

61
[Link]
LOGO
UNIV/POLTEK

Hands-on exercises

 Calculate the basic statistics for the salary column;

 Find how many values in the salary column (use count method);

 Calculate the average salary;

62
[Link]
LOGO
UNIV/POLTEK

Data Frames groupby method

Using "group by" method we can:
• Split the data into groups based on some criteria
• Calculate statistics (or apply a function) to each group
• Similar to dplyr() function in R
In [ ]: #Group data using rank
df_rank = [Link](['rank'])

In [ ]: #Calculate mean value for each numeric column per each group
df_rank.mean()

63
[Link]
LOGO
UNIV/POLTEK

Data Frames groupby method

Once groupby object is create we can calculate various statistics for each group:

In [ ]: #Calculate mean salary for each professor rank:

[Link]('rank')[['salary']].mean()

Note: If single brackets are used to specify the column (e.g. salary), then the output is Pandas Series object.
When double brackets are used the output is a Data Frame 64
[Link]
LOGO
UNIV/POLTEK

Data Frames groupby method

groupby performance notes:

- no grouping/splitting occurs until it's needed. Creating the groupby object
only verifies that you have passed a valid mapping
- by default the group keys are sorted during the groupby operation. You may
want to pass sort=False for potential speedup:

In [ ]: #Calculate mean salary for each professor rank:

[Link](['rank'], sort=False)[['salary']].mean()

65
[Link]
LOGO
UNIV/POLTEK

Data Frame: filtering

To subset the data we can apply Boolean indexing. This indexing is commonly
known as a filter. For example if we want to subset the rows in which the salary
value is greater than $120K:
In [ ]: #Calculate mean salary for each professor rank:
df_sub = df[ df['salary'] > 120000 ]

Any Boolean operator can be used to subset the data:

> greater; >= greater or equal;
< less; <= less or equal;
== equal; != not equal;
In [ ]: #Select only those rows that contain female professors:
df_f = df[ df['sex'] == 'Female' ]
66
[Link]
LOGO
UNIV/POLTEK

Data Frames: Slicing

There are a number of ways to subset the Data Frame:
• one or more columns
• one or more rows
• a subset of rows and columns

Rows and columns can be selected by their position or label

67
[Link]
LOGO
UNIV/POLTEK

Data Frames: Slicing

When selecting one column, it is possible to use single set of brackets, but the
resulting object will be a Series (not a DataFrame):
In [ ]: #Select column salary:
df['salary']

When we need to select more than one column and/or make the output to be a
DataFrame, we should use double brackets:
In [ ]: #Select column salary:
df[['rank','salary']]

68
[Link]
LOGO
UNIV/POLTEK

Data Frames: Selecting rows

If we need to select a range of rows, we can specify the range using ":"

In [ ]: #Select rows by their position:

df[10:20]

Notice that the first row has a position 0, and the last value in the range is omitted:
So for 0:10 range the first 10 rows are returned with the positions starting with 0
and ending with 9

69
[Link]
LOGO
UNIV/POLTEK

Data Frames: method loc

If we need to select a range of rows, using their labels we can use method loc:

In [ ]: #Select rows by their labels:

df_sub.loc[10:20,['rank','sex','salary']]

Out[ ]:

70
[Link]
LOGO
UNIV/POLTEK

Data Frames: method iloc

If we need to select a range of rows and/or columns, using their positions we can
use method iloc:
In [ ]: #Select rows by their labels:
df_sub.iloc[10:20,[0, 3, 4, 5]]

Out[ ]:

71
[Link]
LOGO
UNIV/POLTEK

Data Frames: method iloc (summary)

[Link][0] # First row of a data frame
[Link][i] #(i+1)th row
[Link][-1] # Last row

[Link][:, 0] # First column

[Link][:, -1] # Last column

[Link][0:7] #First 7 rows

[Link][:, 0:2] #First 2 columns
[Link][1:3, 0:2] #Second through third rows and first 2 columns
[Link][[0,5], [1,3]] #1st and 6th rows and 2nd and 4th columns

72
[Link]
LOGO
UNIV/POLTEK

Data Frames: Sorting

We can sort the data by a value in the column. By default the sorting will occur in
ascending order and a new data frame is return.

In [ ]: # Create a new data frame from the original sorted by the column Salary
df_sorted = df.sort_values( by ='service')
df_sorted.head()

Out[ ]:

73
[Link]
LOGO
UNIV/POLTEK

Data Frames: Sorting

We can sort the data using 2 or more columns:
In [ ]: df_sorted = df.sort_values( by =['service', 'salary'], ascending = [True, False])
df_sorted.head(10)

Out[ ]:

74
[Link]
LOGO
UNIV/POLTEK

Missing Values
Missing values are marked as NaN
In [ ]: # Read a dataset with missing values
flights = pd.read_csv("[Link]

In [ ]: # Select the rows that have at least one missing value

flights[[Link]().any(axis=1)].head()

Out[ ]:

75
[Link]
LOGO
UNIV/POLTEK

Missing Values
There are a number of methods to deal with missing values in the data frame:

[Link]() description
dropna() Drop missing observations

dropna(how='all') Drop observations where all cells is NA

dropna(axis=1, how='all') Drop column if all the values are missing

dropna(thresh = 5) Drop rows that contain less than 5 non-missing values

fillna(0) Replace missing values with zeros

isnull() returns True if the value is missing

notnull() Returns True for non-missing values

76
[Link]
LOGO
UNIV/POLTEK

Missing Values
• When summing the data, missing values will be treated as zero
• If all values are missing, the sum will be equal to NaN
• cumsum() and cumprod() methods ignore missing values but preserve them in
the resulting arrays
• Missing values in GroupBy method are excluded (just like in R)
• Many descriptive statistics methods have skipna option to control if missing
data should be excluded . This value is set to True by default (unlike R)

77
[Link]
LOGO
UNIV/POLTEK

Aggregation Functions in Pandas

Aggregation - computing a summary statistic about each group, i.e.
• compute group sums or means
• compute group sizes/counts

Common aggregation functions:

min, max
count, sum, prod
mean, median, mode, mad
std, var

78
[Link]
LOGO
UNIV/POLTEK

Aggregation Functions in Pandas

agg() method are useful when multiple statistics are computed per column:
In [ ]: flights[['dep_delay','arr_delay']].agg(['min','mean','max'])

Out[ ]:

79
[Link]
LOGO
UNIV/POLTEK

Basic Descriptive Statistics

[Link]() description
describe Basic statistics (count, mean, std, min, quantiles, max)

min, max Minimum and maximum values

mean, median, mode Arithmetic average, median and mode

var, std Variance and standard deviation

sem Standard error of mean

skew Sample skewness

kurt kurtosis

80
[Link]
LOGO
UNIV/POLTEK

Graphics to explore the data

Seaborn package is built on matplotlib but provides high level
interface for drawing attractive statistical graphics, similar to ggplot2
library in R. It specifically targets statistical data visualization

To show graphs within Python notebook include inline directive:

In [ ]: %matplotlib inline

81
[Link]
LOGO
UNIV/POLTEK

Graphics
description
distplot histogram

barplot estimate of central tendency for a numeric variable

violinplot similar to boxplot, also shows the probability density of the
data
jointplot Scatterplot

regplot Regression plot

pairplot Pairplot

boxplot boxplot

swarmplot categorical scatterplot

factorplot General categorical plot

82
[Link]
LOGO
UNIV/POLTEK

Basic statistical Analysis

statsmodel and scikit-learn - both have a number of function for statistical analysis

The first one is mostly used for regular analysis using R style formulas, while scikit-learn is
more tailored for Machine Learning.

statsmodels:
• linear regressions
• ANOVA tests
• hypothesis testings
• many more ...

scikit-learn:
• kmeans
• support vector machines
• random forests
• many more ...

See examples in the Tutorial Notebook

83
[Link]
LOGO
UNIV/POLTEK

Follow our social

media!

[Link]
[Link]
DTS_kominfo
Digital Talent Scholarship 2019

Pusat Pengembangan Profesi dan Sertifikasi

Badan Penelitian dan Pengembangan SDM
Kementerian Komunikasi dan Informatika
Jl. Medan Merdeka Barat No. 9
(Gd. Belakang Lt. 4 - 5)
Jakarta Pusat, 10110

[Link] 84
[Link]