15.6 Accessing Elements from DataFrame- iloc

This video lesson is available to enrolled students.

Enroll to watch — BPSC TRE 4.0

AI Summary

An AI-generated summary of this video lecture.

This video is a tutorial on accessing elements in a Pandas DataFrame using the iloc method. The instructor begins by introducing iloc as an Integer Position Based Indexing method that uses only position numbers (0, 1, 2, ...). He then demonstrates its application with a practical example, creating a DataFrame from a dictionary of data containing names, ages, and marks. The tutorial progresses through several use cases: accessing a single row by its position (df.iloc[0]), accessing a single column by its position (df.iloc[:, 2]), accessing a specific cell by row and column position (df.iloc[2, 1]), accessing multiple rows (df.iloc[1:3]), and accessing multiple columns (df.iloc[:, 0:2]). The instructor uses a digital whiteboard to write code and draw diagrams, explaining the syntax and the concept of exclusive end ranges for slicing.

Chapters

1. 0:00 – 2:00 00:00-02:00

   The video starts with a title slide, "Access Elements from DataFrame : iloc". The instructor introduces the iloc method, defining it as "Integer Position Based Indexing" and stating that it uses only position numbers (0, 1, 2, ...). He explains that iloc is used to access data based on the integer position of rows and columns, not their labels. The instructor then transitions to a practical example, writing Python code to import pandas and create a DataFrame named 'df' from a dictionary of data. The dictionary contains lists for 'Name', 'Age', and 'Marks', and the DataFrame is created with a custom index ['S1', 'S2', 'S3', 'S4']. The resulting DataFrame is displayed on the screen, showing four rows of student data.

2. 2:00 – 4:45 02:00-04:45

   The instructor demonstrates accessing a single row using iloc. He writes the code `print(df.iloc[0])` and explains that this retrieves the first row (position 0) of the DataFrame, which corresponds to the student 'Amit'. He then shows how to access a single column by writing `print(df.iloc[:, 2])`, which selects all rows (indicated by `:`) and the column at position 2 (the 'Marks' column), resulting in a Series of marks. Next, he demonstrates accessing a specific cell with `print(df.iloc[2, 1])`, which retrieves the value at row 2 and column 1, resulting in the age '21'. He then covers accessing multiple rows with `print(df.iloc[1:3])`, explaining that the end index is exclusive, so it returns rows at positions 1 and 2. Finally, he shows how to access multiple columns with `print(df.iloc[:, 0:2])`, which selects all rows and columns from position 0 to 1 (exclusive of 2), resulting in a DataFrame with only the 'Name' and 'Age' columns. Throughout, he uses a digital pen to draw arrows and circles on the screen to visually connect the code to the data.

The video provides a comprehensive, step-by-step guide to the iloc method in Pandas. It begins with a clear definition and then systematically builds understanding through a series of practical examples. The instructor progresses from simple single-element access to more complex multi-element slicing, consistently reinforcing the core concept that iloc operates on integer positions. The use of a digital whiteboard to write code and draw diagrams effectively illustrates the relationship between the syntax and the data, making the abstract concept of positional indexing concrete and easy to understand.