Python Pandas for Business Analytics/ Data Science _ Level 1

In this lecture, we have discussed:

1. Downloading Python

2. Downloading & Installing Anaconda Distribution (which is considered as the birthplace of Python data science)

3. Advantages of using Anaconda Distribution, and available IDEs of Python.

In this lecture, we have discussed:

1. Various options to run Python code _ Command-Line, IDEs, and Code Editors

2. Advantages (& Limitations) of Google Colaboratory [popularly known as ‘Colab’]

3. Getting started with Google Colab

4. Some Amazing Features of Colab

Note: You can use any Python IDE of your choice for the course.

In this lecture, we have discussed:

1. The difference between Python files (.py files) Vs Ipython notebooks (.ipynb files)

2. Getting started with Jupyter & Jupyter Labs

3. How Jupyter lab is different from Jupyter Notebook? and why should you start using Jupyter Lab?

In this lecture, we have discussed:

1. Downloading & installing Spyder (A very popular python IDE)

2. Various plugins available, Updating or installing a specific version

3. Exploring Spyder (without installing it) directly in your browser using Binder, Introduction to its interface, & some short-cuts.

This video will really help you decide whether you want to work in Spyder or not.

External Resources:

Spyder HomePage: https://www.spyder-ide.org/

A quick tour, installation guide, FAQs, and much more: https://docs.spyder-ide.org/current/index.html

In this lecture, we have discussed the process of changing the default startup folder for the Jupyter notebook using anaconda.

Code to be written in the command prompt:

Jupyter notebook --generate-config

In this lecture, we have discussed two commonly used Pandas data structures _Series & Data Frame.

By the end of this lecture, you will gain ??:

1. The basic understanding of the Series & DataFrame & how they are different from each other.

2. Various ways to create a Pandas Series.

3. Series Methods

4. Series Slicing

5. For absolute beginners, we've also briefly discussed Data Types, Data Structure in Python & Typecasting

? More Resources:
Python Data Type: https://youtu.be/C76obkbUGaI
Typecasting in Python: https://youtu.be/FJ6IkFycCdA
Python Basics Playlist: https://youtube.com/playlist?list=PLfZXabiF_17rjR9OMq1jD2zyWVsOOExy0

NOTE ?: Pandas panel data structure is deprecated since version 0.20.0. Refer: https://pandas.pydata.org/pandas-docs/version/0.23/generated/pandas.Panel.html

Usually, we import external data to analyze, however many times, we might need to create our own tabular data [like data frame].
In this lecture, we have discussed different ways to create Pandas Data Frame.

By the end of this lecture, you will be able to create a data frame using ??:

• List of lists

• List of dictionaries

• Dictionary of lists

• Dictionary of series

• List of series

⚡ It might be overwhelming for beginners, we've provided a separate lecture to summarise this again. Do not forget to watch that video for a better understanding!

In this lecture, we have discussed data import & export in Python Pandas along with various ways to load in-built datasets in Python.

1. Importing & Exporting text files.

2. Importing & Exporting CSV files.

3. Exporting & Exporting Excel files [A particular sheet, multiple sheets, or entire workbook]

4. Importing & Exporting HTML files

In this lecture, we’ve demonstrated doing all these operations using Google Collaboratory. There are various ways to get data in Colab, which we’ve discussed:

1. From a web URL

2. From Local Machine

You can also import data sets from your Google Drive or GitHub repository.

Also, we’ve discussed loading in-built data sets from popular python libraries such as statsmodel, seaborn, pydataset, sklearn, etc., and ways to access R data sets too [For those who are familiar with in-built data sets in R]. We’ve also discussed using the ‘chunk-size’ option for larger data files.

Additionally, we’ve provided source code for import & export data sets in Python IDEs other than Colab. For Example, if you’re using Jupyter Notebook, you need to take care of the working directory & absolute/relative path. You might need to install a few libraries too [which come by default with Google Colab]. All these points are explained in great detail in the attached notebook.

⚡ I would recommend pausing the video & practicing along.

In this lecture, we have discussed various ways to filer Data Frames in Python Pandas.

1. Data Structure in Python and Data Structure in Pandas

2. What is Data Filtering & Why do we need it?

3. Data Frame Slicing in Pandas

• Selecting Column(s)

        o By Data Types

        o By Name(s)

        o By index using iloc

        o Based on one or more conditions using loc

• Selecting Rows(s)

       o By index using iloc

       o By Name(s)

       o Based on one or more conditions using loc

• Selecting both Row(s) and Column(s)

      o By index using iloc

      o By Name(s)

      o Based on one or more conditions using loc

In this lecture, we have discussed various ways to explore Data Frames in Python Pandas.

1. Shape, Size & Dimension of Data

2. Row/Column Names, Resetting Index, etc.

3. Statistical Summaries & Aggregation

4. Methods and Cross-Tabulation for Non-numeric variables

5. Data Manipulation & Creating a new column

In this lecture, we have discussed how to create quick plots in Python Pandas. After practicing this lesson, you'll learn:

1. What should you know to select the appropriate chart?

2. When to use which chart? [For more details, check-out additional resources]

3. Creating commonly used charts such as Histogram, Boxplot, Bar Chart, Line Chart, and Scatter Plot, etc.

4. Creating Subplots

5. Customizing Plots

After completing this lesson, you’ll learn:

1. What’s data cleaning?

2. How do you decide what to do with missing values?

3. How to treat missing values using Python Pandas?

In this lecture, we have discussed various ways to combine Data Frames in Python Pandas.

1. Left, Right, Inner & Outer Joins

2. Their similarity with set-theoretic operations & SQL joins

3. pd.merge() Vs df.merge() functions

4. Important Arguments

5. One-to-One and Many-to-One Joins

6. Quick Summary of the lesson

7. Additional Resources

In this lecture, we have discussed some more ways to combine Data Frames in Python Pandas.

1. Row-wise & Column wise Concatenation using pd.concat()

2. Inner & Outer Joins

3. Concatenating multiple Data Frames

4. Using pd.append() to combine DataFrames row-wise

5. A quick Summary highlighting differences between join, merge, concat & append functions

Practice Assignment with solution

Practice Assignment for Automobile data