Introduction to Python by WeCloudData

Python for Data Science
Introduction to Python
WeCloudData
@WeCloudData @WeCloudData tordatascience
weclouddata
WeCloudData tordatascience

Data Jobs in Canada
Skills – Data Scientist

Data Jobs in Canada
Skills – Data Analyst

Data Jobs in Canada
Skills – Data Engineer

Data Jobs in Canada
Skills – ML Engineer

Resources
Python
Coding Practice
Coding & Interviews
• LeetCode
• HackerRank
Book Statistics Online Courses
Udemy
• Complete Python Bootcamp
Datacamp
• Introduction to Python

Data Science
Importance of foundations
Data Science
Machine
Learning
Big Data
Data
Engineering
Deep
Learning
ML
Engineering
Focus on one programming language at a time
• Get good at it
Must have skills
• Python
• SQL

Learn Data Science
Follow a Learning Path
Prerequisites
Data Science
Learning Path
• ML algorithms
• 2 Projects
• Interview Practice
Applied ML
• Data wrangling
• Data Visualization
• Predictive Modeling
Data Science
w/ Python
• Big data tools
• ML at scale
• ML deployment
• Job referrals
Big Data
Python
Foundation
SQL for
Data Science

Scientific Python
The ecosystem

Fraud Detection Project Reference Architecture
data lake
Learn Data Science
Understand Data pipelines
data ingestion preprocessing machine learning
real-time scoring
logging and monitoring
reporting & exploration
metadata

Targeting
Profiles
Personaliz
ation
Parse/Filter Classify Synthesize
POI Prediction
Ontology/TaxonomyContexts
URL Parsing
POI Database
Context Extraction
Topic Modeling
Content Classification
Location Classify
Signal Aggregation
Taste Merging
Taste Scoring
Data Data Science Pipelines Data Product
POI Context Builder
Rule-based Predictor
ML Predictor
Location Attributes
Home/Work Predictor
Co-location Location
Graph
• sklearn
• gensim
• nltk
• mrjob
• PySpark
• PySpark
Why Python?
Python in a data science project

Python Data Management
Structured Data with Pandas DataFrame
Row Index Population Area
California 423967 38332521
Florida 170312 19552860
Illinois 149995 12882135
New York 141297 19651127
Texas 695662 26448193
Column
DataFrame
Values
Column
Row
Row
Row
Row
Row
Row Index
Row Index
Row Index
Row Index
Row Index
Column Index Column Index
# row access returns Series
states.loc['Florida']
# column access returns Series
states['area']
California 38332521
Florida 19552860
Illinois 12882135
New York 19651127
Texas 26448193
population 170312
area 19552860
# index based selection
states.iloc[1:3, :1]
Row
Index
Population
Florida 170312
Illinois 149995
Series
Series
DataFrame

Pandas - GroupBy()
City
Ticket
Sales
Toronto 100
Montreal 50
Toronto 20
Halifax 40
Montreal 30
Halifax 60
City
Ticket
Sales
Toronto 100
Toronto 20
City
Ticket
Sales
Montreal 50
Montreal 30
City
Ticket
Sales
Halifax 40
Halifax 60
City
Ticket
Sales
Toronto 60
City
Ticket
Sales
Montreal 40
City
Ticket
Sales
Halifax 50
City
Ticket
Sales
Toronto 60
Montreal 40
Halifax 50
Input
DataFrame
Split
DataFrameGroupBy
Apply (sum)
DataFrameGroupBy
Combine
DataFrame
df = pd.DataFrame({'city' : ['Toronto', 'Montreal', 'Toronto', 'Halifax',
'Montreal', 'Halifax'],
'sales' : [100, 50, 20, 40, 30, 60]})

Pandas - Join/Merge # 1-to-1 join
pd.merge(employee, hr, how='inner', on='employee')
Other features
• Pivot Tables
• Window Functions

DataVisualizations
Matplotlib, Seaborn, Plotly, Bokeh

Python Data Munging
Database Integration

from mrjob.job import MRJob
class MRWordCount(MRJob):
def mapper(self, _, line):
for word in line.split():
yield(word, 1)
def reducer(self, word, counts):
yield(word, sum(counts))
if __name__ == '__main__':
MRWordCount.run()
from pyspark import SparkContext
if __name__ == "__main__":
if len(sys.argv) != 2:
print("Usage: wordcount <file>", file=sys.stderr)
exit(-1)
sc = SparkContext(appName="PythonWordCount")
lines = sc.textFile(sys.argv[1], 1)
counts = lines.flatMap(lambda x: x.split(' '))
.map(lambda x: (x, 1))
.reduceByKey(add)
output = sorted(counts.collect(), key=lambda x: x[1], reverse=True)[:50]
for (word, count) in output:
print("%s: %i" % (word, count))
sc.stop()
API Support for Big Data Platforms
Hadoop/Spark

Machine Learning
Sklearn | Gensim | NLTK

Data Engineering
Data Pipelines with Airflow

Deep Learning
Strong Python Support

Data Types and Representations

Data Representation
Images – Numpy Arrays
import matplotlib.pyplot as plt
%matplotlib inline f =
open("../data/mnist_train_100.csv", 'r’)
mnist = f.readlines() f.close()
# Use matplotlib to visualize the digits
f = plt.figure(figsize=(10,10))
count=0
for line in mnist[:9]:
count += 1
linebits = line.split(',’)
imarray = np.asfarray(linebits[1:]).reshape((28,28))
plt.subplot(3,3,count)
plt.subplots_adjust(hspace=0.5)
plt.title("Label is " + linebits[0])
plt.imshow(imarray, cmap='Greys',
interpolation='None')

Data Representation
Images – Numpy Arrays
from IPython.core.display import display, HTML
display(HTML("<style>.container { width:120% !important; }</style>"))
np.get_printoptions()['linewidth’]
np.set_printoptions(linewidth=200)
digit = np.asfarray(mnist[0].split(',')[1:]).reshape((28,28))
print(digit)

Data Representation
Flat Files
Toronto 2019-11-01 00:00:00 100
Montreal 2019-11-01 00:00:00 50
Toronto 2019-11-02 00:00:00 20
Halifax 2019-11-01 00:00:00 40
Montreal 2019-11-02 00:00:00 30
Halifax 2019-11-02 00:00:00 60
[Toronto , 2019-11-01 00:00:00 , 100]
[Montreal , 2019-11-01 00:00:00 , 50]
[Toronto , 2019-11-02 00:00:00 , 20]
[Halifax , 2019-11-01 00:00:00 , 40]
[Montreal , 2019-11-02 00:00:00 , 30]
[Halifax , 2019-11-02 00:00:00 , 60]
[[Toronto , 2019-11-01 00:00:00 , 100],
[Montreal , 2019-11-01 00:00:00 , 50],
[Toronto , 2019-11-02 00:00:00 , 20],
[Halifax , 2019-11-01 00:00:00 , 40],
[Montreal , 2019-11-02 00:00:00 , 30],
[Halifax , 2019-11-02 00:00:00 , 60]]
rows of string rows of list List of lists

Data Representation
Flat Files with Schema
Toronto 2019-11-01 00:00:00 100
Montreal 2019-11-01 00:00:00 50
Toronto 2019-11-02 00:00:00 20
Halifax 2019-11-01 00:00:00 40
Montreal 2019-11-02 00:00:00 30
Halifax 2019-11-02 00:00:00 60
{City:Toronto , Datetime: 2019-11-01 00:00:00 , Sales: 100}
{City:Montreal , Datetime: 2019-11-01 00:00:00 , Sales: 50}
{City:Toronto , Datetime: 2019-11-02 00:00:00 , Sales: 20}
{City:Halifax , Datetime: 2019-11-01 00:00:00 , Sales: 40}
{City:Montreal , Datetime: 2019-11-02 00:00:00 , Sales: 30}
{City:Halifax , Datetime: 2019-11-02 00:00:00 , Sales: 60}
[{City: Toronto ,
Datetime: 2019-11-01 00:00:00 ,
Sales: 100} ,
{City: Montreal ,
Datetime: 2019-11-01 00:00:00 ,
Sales: 50} ,
{City: Toronto ,
Datetime: 2019-11-02 00:00:00 ,
Sales: 20} ,
{City: Halifax ,
Datetime: 2019-11-01 00:00:00 ,
Sales: 40} ,
{City: Montreal ,
Datetime: 2019-11-02 00:00:00 ,
Sales: 100} ,
{City: Halifax ,
Datetime: 2019-11-02 00:00:00 ,
Sales: 100} ,
]
rows of string
rows of dictionary
List of dictionaries

Data Representation
Tabular data
City Date
Ticket
Sales
Toronto 2019-11-01 00:00:00 100
Montreal 2019-11-01 00:00:00 50
Toronto 2019-11-02 00:00:00 20
Halifax 2019-11-01 00:00:00 40
Montreal 2019-11-02 00:00:00 30
Halifax 2019-11-02 00:00:00 60
Toronto 2019-11-01 00:00:00 100
Montreal 2019-11-01 00:00:00 50
Toronto 2019-11-02 00:00:00 20
Halifax 2019-11-01 00:00:00 40
Montreal 2019-11-02 00:00:00 30
Halifax 2019-11-02 00:00:00 60
rows of string structured table

Pandas
Series
0 0.019111
1 -0.406886
2 -0.684147
3 0.713178
4 -0.573002
array([ 0.01911064, -0.40688553, -0.68414699, 0.71317766, -0.5730021)
Index
Values
array([ 0, 1, 2, 3, 4)
import pandas as pd
from numpy.random import randn
# create a pandas series with random numbers
s1 = pd.Series(randn(5))
print(s1)

Pandas
DataFrame
California 423967
Florida 170312
Illinois 149995
New York 141297
Texas 695662
California 38332521
Florida 19552860
Illinois 12882135
New York 19651127
Texas 26448193
# create the population series
population_dict = {'California': 38332521,
'Texas': 26448193,
'New York': 19651127,
'Florida': 19552860,
'Illinois': 12882135}
population = pd.Series(population_dict)
# create the area series
area_dict = {'California': 423967,
'Texas': 695662,
'New York': 141297,
'Florida': 170312,
'Illinois': 149995}
area = pd.Series(area_dict)
# create the dataframe from a dict of series
states = pd.DataFrame({'population': population,
'area': area})
statesIndex Values
Index Values
# create the dataframe from a dict of series
states = pd.DataFrame({'population': ,
'area':
})
Series
Series
Row
Index
Population Area
California 423967 38332521
Florida 170312 19552860
Illinois 149995 12882135
New York 141297 19651127
Texas 695662 26448193
Columns
Indices
Row
Index
Values

Pandas
DataFrame – Row/Column
Row Index Population Area
California 423967 38332521
Florida 170312 19552860
Illinois 149995 12882135
New York 141297 19651127
Texas 695662 26448193
Column
DataFrame
Values
Column
Row
Row
Row
Row
Row
Row Index
Row Index
Row Index
Row Index
Row Index
Column Index Column Index

Tabular Data Manipulation
Selection/Indexing
City Datetime
Ticket
Sales
Toronto 2019-11-01 00:00:00 100
Montreal 2019-11-01 00:00:00 50
Toronto 2019-11-02 00:00:00 20
Halifax 2019-11-01 00:00:00 40
Montreal 2019-11-02 00:00:00 30
Halifax 2019-11-02 00:00:00 60
City
Ticket
Sales
Toronto 100
Montreal 50
Toronto 20
Halifax 40
Montreal 30
Halifax 60

Filtering
City Datetime
Ticket
Sales
Toronto 2019-11-01 00:00:00 100
Montreal 2019-11-01 00:00:00 50
Toronto 2019-11-02 00:00:00 20
Halifax 2019-11-01 00:00:00 40
Montreal 2019-11-02 00:00:00 30
Halifax 2019-11-02 00:00:00 60
City Date
Ticket
Sales
Halifax 2019-11-01 00:00:00 40
Halifax 2019-11-02 00:00:00 60

Transformation
Cityc Datetime
Ticket
Sales
Toronto 2019-11-01 00:00:00 100
Montreal 2019-11-01 00:00:00 50
Toronto 2019-11-02 00:00:00 20
Halifax 2019-11-01 00:00:00 40
Montreal 2019-11-02 00:00:00 30
Halifax 2019-11-02 00:00:00 60
City Date
Ticket
Sales
Toronto 2019-11-01 100
Montreal 2019-11-01 50
Toronto 2019-11-02 20
Halifax 2019-11-01 40
Montreal 2019-11-02 30
Halifax 2019-11-02 60
Weekday
Monday
Monday
Tuesday
Monday
Tuesday
Tuesday

Data Aggregation
City
Ticket
Sales
Toronto 100
Montreal 50
Toronto 20
Halifax 40
Montreal 30
Halifax 60
City
Ticket
Sales
Toronto 100
Toronto 20
City
Ticket
Sales
Montreal 50
Montreal 30
City
Ticket
Sales
Halifax 40
Halifax 60
City
Ticket
Sales
Toronto 60
City
Ticket
Sales
Montreal 40
City
Ticket
Sales
Halifax 50
City
Ticket
Sales
Toronto 60
Montreal 40
Halifax 50
Input
DataFrame
Split
DataFrameGroupBy
Apply (sum)
DataFrameGroupBy
Combine
DataFrame

GroupBy - aggregate(), agg()
City
Ticket
Sales
Toronto 100
Montreal 50
Toronto 20
Halifax 40
Montreal 30
Halifax 60
City
Ticket
Sales
Toronto 100
Toronto 20
City
Ticket
Sales
Montreal 50
Montreal 30
City
Ticket
Sales
Halifax 40
Halifax 60
City
Ticket
Sales
Toronto 60
City
Ticket
Sales
Montreal 40
City
Ticket
Sales
Halifax 50
City
Ticket
Sales
Toronto 60
Montreal 40
Halifax 50
df.groupby('city').aggregate({'sales': 'mean'})

A filtering operation allows you to drop data
based on the group properties
GroupBy – filter()
City
Ticket
Sales
Toronto 100
Montreal 50
Toronto 20
Halifax 40
Montreal 30
Halifax 60
City
Ticket
Sales
Toronto 100
Toronto 20
City
Ticket
Sales
Montreal 50
Montreal 30
City
Ticket
Sales
Halifax 40
Halifax 60
City
Ticket
Sales
Avg()
Toronto 100 60
Toronto 20 60
City
Ticket
Sales
Avg()
Montreal 50 40
Montreal 30 40
City
Ticket
Sales
Avg()
Halifax 40 50
Halifax 60 50
City
Ticket
Sales
Toronto 100
Toronto 20
Halifax 40
Halifax 60
df.groupby('city').filter(lambda x: x['sales'].mean()>45)

GroupBy – transform()
While aggregation must return a reduced version of the data, transformation can return some transformed version of the full data
to recombine. For such a transformation, the output is the same shape as the input
City
Ticket
Sales
Toronto 100
Montreal 50
Toronto 20
Halifax 40
Montreal 30
Halifax 60
City
Ticket
Sales
Toronto 100
Toronto 20
City
Ticket
Sales
Montreal 50
Montreal 30
City
Ticket
Sales
Halifax 40
Halifax 60
City
Ticket
Sales
Sum()
Toronto 100 120
Toronto 20 120
City
Ticket
Sales
Sum()
Montreal 50 80
Montreal 30 80
City
Ticket
Sales
Sum()
Halifax 40 100
Halifax 60 100
City
Ticket
Sales
Toronto 0.833
Toronto 0.167
Montreal 0.625
Montreal 0.375
Halifax 0.4
Halifax 0.6
df.groupby('city').transform(lambda x: x / x.sum())

• INNER JOIN: Select records that have matching values in both tables.
• LEFT OUTER JOIN: Select records from the first (left-most) table with matching
right table records.
• RIGHT OUTER JOIN: Select records from the second (right-most) table with
matching left table records.
• FULL OUTER JOIN: Selects all records that match either left or right table records.
Types of Data Transformation
Join/Merge

Left Right
Inner Join

Left Right
Left Join

Left Right
Full Outer Join

Learning Python
The 3 stages of transitions

1. Start the Anaconda
Navigator
2. Click the Spyder launch
button
3. For this course we will be
using the Jupyter
Notebook
Python IDE
Anaconda

Jupyter Notebook - Local Server/Laptop
Browser
Notebook Server
Python
R
Julia
Kernels
Jupyter Notebook
Local Server/Laptop

Browser
NotebookServer
Python
R
Julia
Kernels
Jupyter Notebook
Remote Server

Jupyter Notebook
Start Jupyter from the Command Line

notebook
running on
port 8892
notebook running on port
8892
Jupyter Notebook
Ports

Learn Python and Data Science with WeCloudData

Data Science with Python
Student Project Demo

Analyzing the Top Travel Influencers on Instagram

Open a webpage
and get HTML of
search results page
Fishing Online
Process data using
pandas and save data
into .csv file
Locating page
elements by XPath
Extract Target Data
NextURL?
Yes
No

Learn Data Science
Understand the big picture

Data Science Fundamentals
Python Programming
WeCloudData

Python Programming
Why Python?
• Python is the most popular data
science and AI programming
language
• Many employers prefer candidates
with python skills
• Mastering python will expose you
to not only Data Scientist jobs, but
also Data Engineers and DevOps

Python Programming
Who is this course for?
• The Python Programming course is designed for students and professionals who want to
learn to code in Python before moving on to the data science and machine learning topics
• Whether you’re considering a new career in data science or upskilling yourself to catch the
latest data science technology trends, this course is designed for you
• If you come from an Excel background, this course teaches you programming fundamentals
and help you gain confidence in coding through implementing projects and use cases
• If you come from SAS or R programming background, this course will help you comprehend
the difference and advantage of Python to other analytics tools
• This course is designed for students who don’t have prior experience with Python

Python Programming
Learning outcome
After completing this course, you will be able to
• Master python programming essentials for data science and analytics
• Use Python to analyze social media data and visualize data using Pandas and Matplotlib
• Understand various business use cases and most importantly how to apply best
practices to real company data problems using python
• Implement your first machine learning algorithm using python via a hands-on project
• Gain real-world experience through hands-on labs and convince your employers that
you’re up for new data analytics projects at work
• Most importantly, you will be able to build your confidence in python coding and get
prepared for learning machine learning and data science

Python Programming
Instructor – Jodie Zhu
• Machine Learning Enthusiast at Dessa
• University of Toronto, Master of Science (Biostatistics)
• Python Instructor at WeCloudData
• Career development mentor
• Expertise: Python | Data Science | Deep Learning
Machine Learning Engineer
Dessa

Python Programming
Prerequisites
Prerequisites
• There are no requirements other than a working laptop and internet connection.
• You do not need prior experience with Python or another programming language such
as R, SAS, or Matlab, but it helps!
Software Requirements
• Anaconda Distribution
• (Installation guidance will be shared after students register for the course. Our assistant
instructor will make sure you have the right software installed for the course

Python Programming
Syllabus
• Python use cases
• Branching
• Loops
• Data Types: list, tuple,
set
• Functions
• Lab: social media
analytics with python
Day 1
Python Basics
Day 2
Intermediate Python
Day 3
Python Data Analysis
• Data Types: String
• Data Types: Dictionary
• Comprehensions
• Regular expression
• Modules & Packages
• Class and Object
• Interview – Prepare for
Python Interview Tests
• Lab – Class and object
• Pandas introduction
• Intro to visualization
with python
• Accessing database
with python
• Use case: Python for
ML and AI
• Project: Building your
first ML algorithm with
python
• Python Installation
• Jupyter Introduction
• Python Introduction
• DS Introduction
• Twitter Dev API Setup
Pre-course
Installation & Preview
Python learning resources, pdf, coding
interview questions

Python Programming
Industry Use Cases
In this course, we not only teach students how to program Python, but also real-world
business use cases. Understanding the use cases and best practices will allow the students
to confidently apply Python skills to practical data challenges
Business Use Cases
• Why is Python programming important for data science, AI, and Cloud?
• Python use cases in Data Science
• Python use cases in AI
• Python use cases in the financial industry
• Python use cases in sales analytics
• Python use cases in social media analytics

Python Fundamentals
Learning Support
Support you will receive during this course include
• Mentorship and advice from industry experts
• In-classroom learning assistance by our assistant instructor
• Online learning support via Slack from instructor and TA
• Hands-on labs and projects to help you apply what you learn
• Additional resources to help you gain advanced knowledge
• Help from our learning advisor on how to choose the learning path and
specialization courses after the Python Programming course

Python Fundamentals
Testimonials

Python Fundamentals
How to convince your employer
Do you know that most employers will reimburse the training costs?
• We have a detailed course syllabus and email template that you can use to convince
your manager that this is the right course for you and a good investment for your
company
• You will have a completed project and presentation that you can use to demo to
your manager and showcase your newly minted Python skills and get ready for more
interesting data analytics projects

Python Fundamentals
FAQ
Questions Answers
For this course, do I need to know programming? No. Students don’t need to know Python or other programming
languages before taking this course
Do I need to complete hands-on projects? Yes. You will need to complete the project and pass a quiz
before you’re qualified for continuing the data science courses
How does this Python course compare to other
comparable courses?
It’s instructor-led, hands-on, and project-based. You learn to
get started with Python programming with a fraction of the
cost from our competitors
What is the teaching philosophy? Use-case driven, hands-on and fun!
How many hours a week do I expect to put in? Most labs are completed in the class. Expect to put in
additional 3-4 hours a week to work on projects and
assignments
Do I get career support if I take a part-time course? No, career support is not included in any courses. We suggest
students work with a 3rd party career service provider such as
WeCareer.

Introduction to Python by WeCloudData

Introduction to Python by WeCloudData

Recommended

Recommended

More Related Content

Similar to Introduction to Python by WeCloudData

Similar to Introduction to Python by WeCloudData (20)

More from WeCloudData

More from WeCloudData (16)

Recently uploaded

Recently uploaded (20)

Introduction to Python by WeCloudData