Hands-on Data Analysis and Visualization with Pandas: Engineer, Analyse and Visualize Data, Using Powerful Python Libraries

by PURNA CHANDER RAO. KATHUL

Length: 316 pages
Edition: 1
Language: English
Publisher: BPB Publications
Publication Date: 2020-08-14
ISBN-10: 9389845645
ISBN-13: 9789389845648
Sales Rank: #4701029 (See Top 100 Books)

21 ratings

Print Book Look Inside

“Get familiar with various Supervised, Unsupervised and Reinforcement learning algorithms

Key Features

Understand the types of Machine learning.
Get familiar with different Feature extraction methods.
Get an overview of how Neural Network Algorithms work.
Learn how to implement Decision Trees and Random Forests.
The book not only explains the Classification algorithms but also discusses the deviations/ mathematical modeling.

Description

This book covers important concepts and topics in Machine Learning. It begins with Data Cleansing and presents an overview of Feature Selection. It then talks about training and testing, cross-validation, and Feature Selection. The book covers algorithms and implementations of the most common Feature Selection Techniques. The book then focuses on Linear Regression and Gradient Descent. Some of the important Classification techniques such as K-nearest neighbors, logistic regression, Naïve Bayesian, and Linear Discriminant Analysis are covered in the book. It then gives an overview of Neural Networks and explains the biological background, the limitations of the perceptron, and the backpropagation model. The Support Vector Machines and Kernel methods are also included in the book. It then shows how to implement Decision Trees and Random Forests.

Towards the end, the book gives a brief overview of Unsupervised Learning. Various Feature Extraction techniques, such as Fourier Transform, STFT, and Local Binary patterns, are covered. The book also discusses Principle Component Analysis and its implementation.

What will you learn

Learn how to prepare Data for Machine Learning.
Learn how to implement learning algorithms from scratch.
Use scikit-learn to implement algorithms.

Who this book is for

The book is designed for Undergraduate and Postgraduate Computer Science students and for the professionals who intend to switch to the fascinating world of Machine Learning. This book requires basic know-how of programming fundamentals, Python, in particular.

Cover Page
Title Page
Copyright Page
Dedication Page
About the Author
About the Reviewer
Acknowledgement
Preface
Errata
Table of Contents
1. Introduction to Data Analysis
    Structure
    Objective
    Inspiration for data analysis
        What is data science?
        Structured data
        Semi-structured data
        Unstructured data
    Domain expertise
    Maths and statistics
    Artificial intelligence
    Machine learning
        Supervised learning
        Unsupervised learning
        Reinforcement learning
    Data infrastructure
    Data analysis process
        Business requirements
        Data collection
        Data cleansing
        Data exploring and visualization
        Data modeling
        Model validation and testing
        Deployment
    Why Python for data analysis?
        Growth of Python programming language
        Prototype extension
        Python libraries for data analysis and installation
            JupyterLab
            Pandas
            Numpy
            Matplotlib
            Seaborn
    Conclusion
    Exercises
    Answers
2. JupyterLab
    Structure
    Objective
    Introduction to JupyterLab
        Architecture
    Components
        Console
        Interface
            Creating a new notebook
    Cell modes
        Code
        Markdown
            Headings
            Lists and unordered lists
            Formatting words
            Math equations
            Load images
            Load videos
        Raw
    Menu
    Magic commands
        %bash
        !
        %run
    Keyboard shortcuts
    Conclusion
    Exercises
    Answers
3. Python Overview
    Structure
    Objective
    Python, Hello World
    Variables and data types
        Strings
        Integer and float
        List
        Tuple
        Dictionary
        Sets
    Functions
        Docstrings
            Function arguments
        Required arguments
        Keyword arguments
        Default arguments
        Variable-length arguments
    Lambda
    List comprehensions
    Functional programming using (map, filter, and reduce)
        Map
        Filter
        Reduce
    Working with datetime objects
        Constructing datetime objects
        Timedelta
    Conclusion
    Exercises
    Answers
4. Introduction to Numpy
    Structure
    Objective
    Ndarray
    Difference between List and Numpy arrays
    Storage
    Type check
    Speed
    Copying arrays
    Mathematical operations
    Trigonometric functions
    Statistical operations
    Reshaping
    Vertical and horizontal stacking of Numpy arrays
    Fancy indexing
    Indexing with Boolean arrays
    Broadcasting
    Conclusion
    Exercises
    Answers
5. Introduction to Pandas
    Structure
        Objective
    Data structures in pandas
    Series
        Creating a series object with strings
        Creating a series object with integers
        Creating a series object with booleans
        Creating a series object with tuples
        Creating a series object with a dictionary
        Series attributes
        Series methods
    DataFrames
        Create a dataframe from the dictionary
        Create a dataframe object with a list of lists
        Creating a dataframe object with a list of dictionaries
        DataFrame methods and attributes
            Head
            Tail
            Shape
            Index
            Count
            Columns
    Conclusion
    Exercises
    Answers
6. Data Analysis
    Structure
    Objective
    Handling different file formats
    Handling rows and columns
        Rename column headers
        Select a series from dataframe
        Updating a series object
        Datatype conversion
        Adding new series to a dataframe
        Deleting series objects
        Filtering rows and columns
        Selecting rows and columns using Index
            Loc
            iLoc
    Groupby
        The internal implementation of groupby
            Splitting the Object
            Applying a function
            Combine
        Aggregate
        Transform
    Filter
    Concatenate DataFrames
    Merge DataFrames
        Many to one
        Inner join
        Joins
            Left join
            Right join
            Outer join
    Purging duplicate rows
    Data Transformations
        Apply
        Map
        ApplyMap
    Crosstab
    Cleansing the Data
        Handling missing values
        Find NaN in series and dataframe objects
        Select the missing values
        Drop missing values from Series
        Drop NaN from DataFrame
        Fill missing values
        Forward and backward fill
        Filling the DateTime values
    Replacing individual values
    Pivot and pivot table
        Pivot table
    Grouper
    Handling large datasets
        Optimizing the data types
            Memory usage
            Datatype subtypes
            Optimizing the Numeric columns
            Optimizing the Object columns
        Converting to Pickle file
            Read the file in chunks
    Modin pandas
        Dask
    Conclusion
    Exercises
    Answers
7. Time Series Analysis
    Structure
    Objective
    Creating time series data
        Start and end dates
        Periods
        Frequency weekly, monthly, hourly and seconds
    Converting string-based dates to datetime objects
    Unix / Epoch time
    Time series analysis using a real-time dataset
        Slicing and indexing
        Resampling
        Plotting
    Handling timezones
    Shifting or lagging
    Handling holidays
    Conclusion
    Exercises
    Answers
8. Introduction to Statistics
    Structure
    Objective
    Population
    Sample
    Types of data
        Categorical variables
        Numerical data
    Levels of measurement
        Qualitative nominal
        Qualitative ordinal
        Quantitative interval
            Example
        Quantitative ratio
            Example
        Descriptive statistics
        Measures of central tendency
            Mean
            Median
            Mode
        Measures of variability
            Range
            Quartile
            Deviation/Variance
            Standard deviation
            Coefficient of variation
            Covariance
    Inferential statistics
        What is a distribution?
        Standardization (Z-score)
        Central limit theorem
        Standard error
        Confidence intervals
    Hypothesis Testing
        Null hypothesis
        Alternate hypothesis
        Errors
            Type 1 error
            Type 2 error
        One-tailed test
        Two-tailed test
            Hypothesis testing steps
        The T-Test (Student T-Test)
        Z-Test
    Conclusion
    Exercises
    Answers
9. Matplotlib
    Structure
    Objective
    Why data visualization?
    Matplotlib architecture
        Backend layer
        Artistic layer
        Scripting layer
    Chart properties
        Figure and axes
        Labeling the axes
        Formatting
            Saving the chart
        Adding annotations
        Legends
        Controlling axes limits
    Controlling xticks, y_ticks, and tick_labels
    Scatter plot
    Bar plot
    Histograms
        Line plot
    Pie chart
    Subplots
    Conclusion
    Exercises
    Answers
10. Seaborn
    Structure
    Objective
    Why Seaborn?
    Matplotlib versus Seaborn
        Seaborn chart properties
        Styles
        Axes spines
        Controlling style properties
        Scaling plot elements
        Color palette
            Qualitative
            Sequential
            Diverging
    About pokemon
    Importing libraries and dataset
    Visualizing statistical relationships
    Plotting categorical variables
    Visualizing the distribution of the data
        Univariate
        Bivariate
        Multiplot grids
        Pair grid
    Conclusion
    Exercises
    Answers
11. Exploratory Data Analysis
    Structure
    Objective
    A little story, Titanic
    Importing libraries and dataset
    Handling missing values
    Variable identification
    Categorical nominal
        Categorical ordinal
        Numerical continuous
        Numerical discrete
    Univariate analysis
    Bivariate analysis
        Scatter plot
        Line plot
    HeatMap
        Bar plot
    Multivariate analysis
    Handling outliers
    Feature Selection
    Conclusion
    Exercises
    Answers