apply() on columns in Pandas - Step-by-Step Execution

Concept Flow - apply() on columns

Start with DataFrame

↓

Select a column

↓

Apply function to each value

↓

Collect results into new Series

↓

Replace or add column in DataFrame

↓

End

The apply() method takes a function and runs it on each value in a DataFrame column, then collects the results back into a Series or DataFrame column.

Execution Sample

Pandas

import pandas as pd

df = pd.DataFrame({'A': [1, 2, 3]})
df['B'] = df['A'].apply(lambda x: x * 2)
print(df)

This code doubles each value in column 'A' and stores the result in a new column 'B'.

Execution Table

Step	Input Value (x)	Function Applied	Result	DataFrame State
1	1	x * 2	2	{'A': [1, 2, 3], 'B': [2]}
2	2	x * 2	4	{'A': [1, 2, 3], 'B': [2, 4]}
3	3	x * 2	6	{'A': [1, 2, 3], 'B': [2, 4, 6]}
4	-	-	-	Completed applying function to all rows

💡 All values in column 'A' processed; new column 'B' created with doubled values.

Variable Tracker

Variable	Start	After 1	After 2	After 3	Final
df['A']	[1, 2, 3]	[1, 2, 3]	[1, 2, 3]	[1, 2, 3]	[1, 2, 3]
df['B']	[]	[2]	[2, 4]	[2, 4, 6]	[2, 4, 6]

Key Moments - 2 Insights

Why does apply() run the function on each value separately instead of the whole column at once?

apply() is designed to apply the function element-wise on each value in the column, as shown in execution_table rows 1-3 where each input value is processed individually.

What happens if the function returns a different type or size than expected?

The result will be collected into a Series matching the original column length. If the function returns unexpected types, the new column may have mixed types or cause errors. This is why the function should return one value per input.

Visual Quiz - 3 Questions

Look at the execution table, what is the value of df['B'] after step 2?

A [4]

B [2]

C [2, 4]

D []

💡 Hint

Check the DataFrame State column at step 2 in the execution_table.

At which step does the apply() function finish processing all values?

A Step 4

B Step 3

C Step 2

D Step 1

💡 Hint

Look for the row in execution_table where the DataFrame State says 'Completed applying function to all rows'.

If the lambda function was changed to 'x + 1', what would be the value of df['B'] after step 3?

A [3, 4, 5]

B [2, 3, 4]

C [1, 2, 3]

D [0, 1, 2]

💡 Hint

Add 1 to each original value in df['A'] which are [1, 2, 3].

Concept Snapshot

apply() on columns:
- Use df['col'].apply(func) to run func on each value.
- func receives one value at a time.
- Returns a Series with results.
- Can create or replace columns.
- Useful for element-wise transformations.

Full Transcript

This visual execution shows how pandas apply() works on DataFrame columns. Starting with a DataFrame with column 'A', we apply a function that doubles each value. The function runs on each value one by one, producing results collected into a new column 'B'. The execution table traces each step, showing input, function, result, and DataFrame state. Variable tracker shows how df['B'] grows after each step. Key moments clarify why apply() works element-wise and what happens if function returns unexpected results. The quiz tests understanding of the stepwise changes and effects of changing the function. The snapshot summarizes the key points for quick recall.