[Python] Data Reference and Update in Pandas Series (Index and Attribute Access)

The Pandas Series is a data structure that combines the properties of both lists and dictionaries. Therefore, you can intuitively access individual data or rewrite values using defined indices (labels).

This article explains how to retrieve data by specifying indices, access data using attribute (dot) notation, and how to update data using these methods.

Methods for Referencing and Updating Data

There are two main ways to access elements within a Series:

• Bracket Notation []: Specifying the key (index) like a dictionary. This is the safest and most common method.
• Dot Notation . : Specifying the index name as an attribute of the object. This makes the code concise, but it cannot be used if the index name contains reserved words or spaces.

Below is the implementation code using weekly temperature data as an example.

import pandas as pd

def manage_temperature_data():
    """
    Function to demonstrate referencing and updating data in Pandas Series
    """
    # Create maximum temperature data for weekly weather forecast
    # Specify days of the week in the index
    temperatures = [24.5, 25.0, 23.8, 26.1]
    days = ["Monday", "Tuesday", "Wednesday", "Thursday"]
    
    series_temp = pd.Series(temperatures, index=days, name="Temperature")
    
    print("--- Initial Data ---")
    print(series_temp)
    print("\n")


    # 1. Referencing Data
    print("=== Referencing Data ===")
    
    # Access via Bracket Notation (Dictionary style)
    # Specify the value using the string index as a key
    temp_mon = series_temp["Monday"]
    print(f"Temperature on Monday (Bracket Ref): {temp_mon}")

    # Access via Dot Notation (Attribute style)
    # Write the index name directly like a property
    temp_tue = series_temp.Tuesday
    print(f"Temperature on Tuesday (Dot Ref)   : {temp_tue}")
    print("\n")


    # 2. Updating Data
    print("=== Updating Data ===")
    
    # Overwrite value using Bracket Notation
    # Correcting Wednesday's temperature
    series_temp["Wednesday"] = 28.5
    print(f"Updated Wednesday: {series_temp['Wednesday']}")

    # Overwrite value using Dot Notation
    # Correcting Thursday's temperature
    series_temp.Thursday = 27.0
    print(f"Updated Thursday : {series_temp.Thursday}")

    print("\n--- Data After Update ---")
    print(series_temp)

if __name__ == "__main__":
    manage_temperature_data()

Execution Result

--- Initial Data ---
Monday       24.5
Tuesday      25.0
Wednesday    23.8
Thursday     26.1
Name: Temperature, dtype: float64


=== Referencing Data ===
Temperature on Monday (Bracket Ref): 24.5
Temperature on Tuesday (Dot Ref)   : 25.0


=== Updating Data ===
Updated Wednesday: 28.5
Updated Thursday : 27.0

--- Data After Update ---
Monday       24.5
Tuesday      25.0
Wednesday    28.5
Thursday     27.0
Name: Temperature, dtype: float64

Important Note: Limitations of Dot Notation

Dot notation (series.label) is easy to write and highly readable, but it is not all-purpose. You must use bracket notation (series["label"]) in the following cases:

• When the index name is not a valid Python identifier: For example, if it contains spaces (e.g., "New York") or starts with a number (e.g., "1st_Day").
• When it overlaps with Series method names: If names like sum, max, index, or values are used as indices, using dot notation will refer to the method itself rather than the data.

Basically, it is common to use bracket notation, utilizing dot notation in situations where convenience is needed, such as interactive analysis.