13  Pandas module:

Let’s break down the pandas library in simple terms with analogies.

Pandas as a Data Organizer:

Imagine you have a big box of LEGO bricks. Each LEGO brick is like a piece of data. Now, you want to build something cool with these bricks, but managing them in the box can be messy. Here’s where pandas comes in.

  1. Series - The Single LEGO Stack:

    • A Series is like a single stack of LEGO bricks. It’s organized and labeled. Each brick (data point) has its place, and you can easily refer to them by their position or label.
    import pandas as pd

# Creating a Series
temperatures = pd.Series([25, 28, 24, 30, 22], name='Temperature')

    Just like a stack of LEGO bricks neatly arranged, a Series keeps your data in order.

  2. DataFrame - The LEGO Structure:

    • Now, imagine you want to build something more complex, like a spaceship. A DataFrame is like a structured LEGO creation. It consists of multiple stacks (Series), each representing a specific aspect of your project.
    import pandas as pd

# Creating a DataFrame
data = {'Day': ['Mon', 'Tue', 'Wed', 'Thu', 'Fri'],
        'Temperature': [25, 28, 24, 30, 22]}
weather_df = pd.DataFrame(data)

    In this analogy, your spaceship (DataFrame) has different sections (columns) for the day and temperature, and each section is like a well-organized stack of LEGO bricks (Series).

Let’s look another simple example of creating dataframe:

Certainly, Asad_Pro_Beta! Let’s break down the provided code step by step:

import pandas as pd

# Creating a DataFrame
df = pd.DataFrame([[1000, 'steve', 86],
                   [1001, 'mathew', 91],
                   [1002, 'jose', 72],
                   [1003, 'patty', 69],
                   [1004, 'vin', 88]],
                  columns=['Regd.No', 'Name', 'Marks%'],
                  index=['ID1', 'ID2', 'ID3', 'ID4', 'ID5'])

# Displaying the DataFrame
print(df)

Explanation:

  1. Importing Pandas:
    • import pandas as pd: This line imports the pandas library and gives it the alias pd. This alias is commonly used for brevity.
  2. Creating a DataFrame:
    • pd.DataFrame(...): This creates a DataFrame. The data is provided as a list of lists, where each inner list represents a row of data.
  3. Data Values:
    • The inner lists contain values for ‘Regd.No’ (Registration Number), ‘Name’, and ‘Marks%’ respectively for each student.
  4. Columns:
    • columns=['Regd.No', 'Name', 'Marks%']: This specifies the column names for the DataFrame.
  5. Index:
    • index=['ID1', 'ID2', 'ID3', 'ID4', 'ID5']: This sets custom index values for the DataFrame. Each index corresponds to a row.
  6. Displaying the DataFrame:
    • print(df): This prints the DataFrame to the console.

Resulting DataFrame:

     Regd.No   Name  Marks%
ID1     1000  steve      86
ID2     1001  mathew     91
ID3     1002  jose       72
ID4     1003  patty      69
ID5     1004  vin        88

So, the DataFrame df is a table with columns ‘Regd.No’, ‘Name’, and ‘Marks%’, and custom row indices ‘ID1’ through ‘ID5’, representing information about students, their registration numbers, names, and marks percentage. Each row corresponds to a different student.

Note: From now on i will be using Jupyter Notebook to do data analysis because Jupyter Notebook is made for that purpose. To install Jupyter Notebook go on to this Jupyter Notebook

**Creating a DataFrame and adding some data to it.

# Giving alias to pandas module and calling it pd instead of pandas as you would give a nickname to a person
import pandas as pd

df = pd.DataFrame([[1000,'steve',86],
                       [1001,'mathew',91],
                       [1002,'jose',72],
                       [1003,'patty',69],
                       [1004,'vin',88]], columns = ['Regd.No','Name','Marks%'], index=['ID1','ID2','ID3','ID4','ID5'])

df

Reading the data from a webpage

import pandas as pd

url = 'https://en.wikipedia.org/wiki/Python_(programming_language)'

d = pd.read_html(url)
d[1]

Now we will use the above data to perform various operation on it.

You have to convert the DataFrame into string for regular expression to work on it.

content = d[1].to_string()
content[:1000]

Scenario 1: Return Data types that are mutable

content = d[1].to_string()
pattern = r'\n\d{1,}\s{1,}(.+?)\s{1,}mutable.+'

result = re.findall(pattern, content)
result

Scenario 2: Return those data types that uses the curly braces format

pattern2 = r'\d{1,}\s{1,}(\w+?)\s{1,}.+\{.*\}'
result = re.findall(pattern2, content)
result

Scenario 3: Extract datatype names who is no longer than 4 characters e.g. int, set, dict, list so on.

pattern3 = r'\n\d{1,}\s{1,}(\w{3,4})\s{1,}.+'
result = re.findall(pattern3, content)
result

Scenario 3: return the description of those id who is odd

pattern4 = r'\n(\d*[13579])\s{1,}.+?\s{1,}\w+\s{1,}(.{,10}).+'
result = re.findall(pattern4,content)
result

Scenario 4: return all the data type who’s syntax part contain at least one decimal part

pattern5 = r'\n\d{1,}\s{1,}(\w+).+\s{1,}.+(\d+\.\d+)'
result = re.findall(pattern5, content)
result
Back to top