Python for Data Science
Overview
This course – Python for Data Science is an introductory course meticulously crafted to
empower students with the indispensable skills required for not just data analysis but for
crafting data-driven models that can unlock hidden insights and drive informed decisions.
It starts by exploring Pandas, a powerful library for data manipulation and analysis, enabling
students to effectively manage and process data.
As we venture further, we'll paint vivid stories through data visualization using Matplotlib
and navigate the intricate world of exploratory data analysis (EDA) to unveil the narratives
within datasets.
Beyond that, we'll equip you with the tools to handle missing values and outliers, preserving
the integrity of your data. And as the final stroke, we'll delve into data normalization
techniques and the creative artistry of feature engineering, transforming your data into a
masterpiece optimized for machine learning.
By the end of the course, you'll not only have a solid Python foundation but the creative
prowess to sculpt data into actionable insights. This course also acts as the first
steppingstone for aspiring data scientists. This course is a part of the program titled Fractal
Data Science Professional Certificate.
Requirements/Prerequisites
There are no requirements or prerequisites for the course. To succeed in this course, all you
need to have is a curious and logical mind. While having a basic understanding of
programming would be beneficial, it is not mandatory.
Course Structure
The course consists of the following materials:
• Videos: The lectures are delivered via videos, which are broken into small chunks,
usually between five and seven minutes each.
• Expert talks: Valuable insights and expertise into the intricacies of real-world
project dynamics and processes in a concise 15 to 20-minute presentation by
industry experts.
• Reading: We have made available pdf files of all the reading materials. There can be
some external links or references for further reading.
• Ungraded Practice Quizzes: There will be non-graded short "quiz" questions that will
follow some of the videos to help you gauge your understanding.
Python for Data Science
• Ungraded Assignment: There will be ungraded assignments which will help you with
the hands-on practice for every lesson which will give you enough practice to
attempt the graded assignments.
• Course materials: We have included the Python notebooks and datasets that are
utilized in each module to facilitate hands-on practice and learning and help you
refer to the concepts taught through the course..
• Graded Assignment: There will be graded assignments for submission part of every module.
Course Topics
Week 1: Introduction to Python for Data Science
Introduction, Kickstarting with Python and Jupyter, How Python Solved a Business Problem, The
Business Problem for the Course,
Week 2: Data Wrangling with Python
Getting started with a data set, Dataframe Fundamentals and Indexing, Exploring and Querying on
Columns, Manipulation and Grouping of Data, Date Time Operations in Python, Putting Together
Data from Multiple Sources
Week 3: Exploratory Data Analysis
Descriptive Statistics, Understanding Data Distributions, Relationships Between Features, Enhancing Data
Visualizations, Comparing Two Groups: Hypothesis Testing
Week 4: Data pre-processing
Treating Missing Data, Data Transformation, Understanding Data Normalization and Outlier Detection
Week 5: Feature Engineering
Creating Derived Features in Python, Gen AI in Python
Grading
Your grade in the course will be based solely on the graded labs. Ungraded labs
and practice quizzes – will help. You are free to follow the course without
completing the graded assignments and quizzes, but then you will not receive a
certificate of completion.