Python Pandas Library

 Pandas

• Pandas is a Python library used for working with data sets.
• It has functions for analyzing, cleaning, exploring, manipulating data, and plotting data.
• The name "Pandas" has a reference to both "Panel Data" and "Python Data Analysis" and was created by Wes McKinney in 2008.

Main Features

1. Explorin Data
2. Preprocessing / Cleaning Data 
3. Analyzing Data 
4. Visualization of Data

Coding Editor

• Jupyter
• Google Colab

Pandas Installaltion

pip install pandas

import pandas as pd

Exploring Data

1. Reading Data Frame

df = pd.read_csv('sinovac.csv')

2. head() and tail() function

 The head function will display the first rows, and the tail will be the last rows. 

By default, it shows 5 rows.

3. info()

To display data frames information we can use info() the method.

4. Display the number of rows and columns.

df.shape

5. Display column names

df. columns

6. display one column only

df['col name'].head(3)

7. display multiple  columns

df[['Age', 'Transaction_date', 'Gender']].head(4)

Data Cleaning

1. Delete Columns name

df.drop(['Transaction_ID'], axis=1, inplace=True)

2. Change the Columns name

df. rename(columns={"Transaction_date": "Date", "Gender": "Sex"},

inplace=True)

3.  Remove duplicate

# Display duplicated entries 

df. duplicated().sum()

# dropping ALL duplicate values

df.drop_duplicates(inplace = True)

4. Display missing values information

df.isna().sum().sort_values(ascending=False)

#Delete Nan rows of Job Columns

df.dropna(inplace=True)

Data Analysis

1. df. describe()

 This function will display most of the basic statistical measurements. 

2. df['State_names'].unique()

 Shows all unique values 

3.  df['Gender'].value_counts()

Counts of unique values

# Calculate the percentage of each category

df['Gender'].value_counts(normalize=True)

df['State_names'].value_counts().sort_values(ascending = False).head(20)

4. Sort Values by State_names

df.sort_values(by=['State_names']).head(3)

# Sort Values Amount_spent with descending order

df.sort_values(by=['Amount_spent'], ascending=False).head(3)

5.  nlargest() and nsmallest() functions

We can use nlargest() and nsmallest() functions for displaying largest and smallest values with desired numbers.

df.nlargest(4, 'Amount_spent').head(10)

df.nsmallest(3, 'Age').head(10) 

#6. filtering - Only show Paypal users

condition = df['Payment_method'] == 'PayPal'

df[condition].head(4)

Data Visualization

1 . Line plot

# use plot() method on the data frame

df.plot('year', 'price');

2. Bar plot

df['Employees_status'].value_counts().plot(kind='bar');

For vertical bar:

df['Employees_status'].value_counts().plot(kind='barh');

3 Pie plot

df['Segment'].value_counts().plot( kind='pie');