Row and Column Selection Basics

📊 Selecting Data

In the previous tutorial, we learned how to create DataFrames. Now, we will master the art of slicing and dicing data. We will learn how to select specific Columns, specific Rows, and how to combine both techniques.

---

🔹 1. Column Selection

Selecting columns is the most common operation in Pandas. We use square brackets [] directly on the DataFrame.

A. Selecting a Single Column

To select a single column and keep it as a DataFrame, we use double square brackets.

🧪 Code Example

import pandas as pd

# Setup our sample DataFrame
data = {
    "Name": ["Atul", "Aman", "Harsh", "Rishi", "Devansh"],
    "Class": [11, 39, 47, 56, 76],
    "Status": ["pass", "fail", "pass", "pass", "fail"]
}
table = pd.DataFrame(data)

# Select only the "Name" column
print(table[["Name"]])

📤 Output

      Name
0     Atul
1     Aman
2    Harsh
3    Rishi
4  Devansh

⚠ Important Note: Series vs. DataFrame

• If you use single brackets table["Name"], Pandas returns a Series (a simple list).
• If you use double brackets table[["Name"]], Pandas returns a DataFrame (a table).
• Always use double brackets if you want your output to look like a table.

B. Selecting Multiple Columns

To select multiple columns, we pass a list of column names inside the brackets. This creates a 2D array structure.

🧪 Code Example

# Select "Name" and "Class" columns
print(table[["Name", "Class"]]) 

📤 Output

      Name  Class
0     Atul     11
1     Aman     39
2    Harsh     47
3    Rishi     56
4  Devansh     76

---

🔹 2. Row Selection (The .loc Method)

To select rows, we use the .loc (Location) property. This looks up rows by their Index Label.

A. Selecting a Single Row

Use .loc with double brackets to select a specific row index.

🧪 Code Example

# Select the row at index 1
print(table.loc[[1]]) 

📤 Output

   Name  Class Status
1  Aman     39   fail

B. Selecting Multiple Rows

Pass a list of indices to select multiple specific rows.

🧪 Code Example

# Select rows 0, 1, and 2
print(table.loc[[0, 1, 2]])

📤 Output

    Name  Class Status
0   Atul     11   pass
1   Aman     39   fail
2  Harsh     47   pass

ℹ Note on Indexing:

.loc looks for the Label of the index, not the position. While our rows are numbered 0, 1, 2 by default, if your rows were named "A", "B", "C", you would use .loc[["A"]].

---

🔹 3. Combined Selection (Rows & Columns)

This is the most powerful way to select data. You can pinpoint specific data by defining both the rows and the columns you want.

The syntax follows this pattern:
table.loc[ [Row_Labels] , [Column_Labels] ]

A. Specific Row, Specific Column

Here, we ask for Row 0 and only the Name column.

🧪 Code Example

print(table.loc[[0], ["Name"]]) 

📤 Output

   Name
0  Atul

📝 Explanation

• Pandas looks at Row index 0.
• It then intersects that with the Column "Name".
• It returns the exact cell value but formatted as a DataFrame.

B. Multiple Rows, Multiple Columns

Here, we select a block of data: Rows 0, 1, 2 and Columns Name, Class.

🧪 Code Example

print(table.loc[[0, 1, 2], ["Name", "Class"]]) 

📤 Output

    Name  Class
0   Atul     11
1   Aman     39
2  Harsh     47

📝 Explanation

• First Argument [0, 1, 2]: Filters the DataFrame to keep only these three rows.
• Second Argument ["Name", "Class"]: Filters columns to keep only Name and Class.
• The result is a smaller subset table containing only the requested intersection.

---

📌 Summary

Goal	Syntax
Just Columns	df[["Col1", "Col2"]]
Just Rows	df.loc[[0, 1]]
Both (Intersection)	df.loc[[Rows], [Cols]]