Introduction to Pandas 

    

# Dependency needed to install file 

!pip install xlrd


# Import required library

import pandas as pd

  • After the import command, we now have access to a large number of pre-built classes and functions.


  • This assumes the library is installed; . One way pandas allows you to work with data is a dataframe.


  • Let's go through the process to go from a comma separated values (.csv) file to a dataframe.


  • This variable csv_path stores the path of the .csv, that is used as an argument to the read_csv function.


  • The result is stored in the object df, this is a common short form used for a variable referring to a Pandas dataframe. 


    • 

# Read data from CSV file

csv_path = 'https://s3-api.us-geo.objectstorage.softlayer.net/cf-courses-data/CognitiveClass/PY0101EN/Chapter%204/Datasets/TopSellingAlbums.csv'
df = pd.read_csv(csv_path)

  • We can use the method head() to examine the first five rows of a dataframe: 


    • 




# Print first five rows of the dataframe

df.head()

  • We use the path of the excel file and the function read_excel. The result is a data frame as before: 


    • 

# Read data from Excel File and print the first five rows

xlsx_path = 'https://s3-api.us-geo.objectstorage.softlayer.net/cf-courses-data/CognitiveClass/PY0101EN/Chapter%204/Datasets/TopSellingAlbums.xlsx'

df = pd.read_excel(xlsx_path)
df.head()

  • We can access the column Length and assign it a new dataframe x: 


    • 




# Access to the column Length

x = df[['Length']]
x

  • The process is shown in the figure:


  • Read data from Excel File and print the first five rows in pandas


  • Viewing Data and Accessing Data You can also get a column as a series. You can think of a Pandas series as a 1-D dataframe. Just use one bracket: 


    • 


# Get the column as a series

x = df['Length']
x

  • You can also get a column as a dataframe. For example, we can assign the column Artist: 


    • 



# Get the column as a dataframe

x = type(df[['Artist']])
x

  • You can do the same thing for multiple columns; we just put the dataframe name, in this case, df, and the name of the multiple column headers enclosed in double brackets. The result is a new dataframe comprised of the specified columns: 


    • 



# Access to multiple columns

y = df[['Artist','Length','Genre']]
y

  • The process is shown in the figure:


  • Access to multiple columns


  • One way to access unique elements is the iloc method, where you can access the 1st row and the 1st column as follows: 


    • 


# Access the value on the first row and the first column

df.iloc[0, 0]

  • You can access the 2nd row and the 1st column as follows: 


    • 




# Access the value on the second row and the first column

df.iloc[1,0]

  • You can access the 1st row and the 3rd column as follows: 


    • 


# Access the value on the first row and the third column

df.iloc[0,2]

  • You can access the column using the name as well, the following are the same as above: 


    • 


# Access the column using the name

df.loc[0, 'Artist']



 

    

# Access the column using the name

df.loc[1, 'Artist']


 

    


# Access the column using the name

df.loc[0, 'Released']



 

    

# Access the column using the name

df.loc[1, 'Released']

  • You can perform slicing using both the index and the name of the column: 


    • 



# Slicing the dataframe

df.iloc[0:2, 0:3]

 

    



# Slicing the dataframe using name

df.loc[0:2, 'Artist':'Released']