Combining multiple cleaning steps in Pandas - Time & Space Complexity

Time Complexity: Combining multiple cleaning steps

O(n)

Understanding Time Complexity

We want to understand how the time needed changes when we combine several cleaning steps in pandas.

How does the total work grow as the data gets bigger?

Scenario Under Consideration

Analyze the time complexity of the following code snippet.

import pandas as pd

df = pd.DataFrame({
    'A': ['foo', 'bar', None, 'baz', 'foo'],
    'B': [1, 2, 3, None, 5],
    'C': ['x', 'y', 'z', 'x', None]
})

df = df.dropna()
df['A'] = df['A'].str.upper()
df['B'] = df['B'] * 10

This code drops rows with missing values, changes column A to uppercase, and multiplies column B by 10.

Identify Repeating Operations

Identify the loops, recursion, array traversals that repeat.

• Primary operation: Each cleaning step processes the entire DataFrame or columns once.
• How many times: Three main passes: one for dropna, one for uppercasing, one for multiplying.

How Execution Grows With Input

Each step looks at all rows, so the total work grows roughly in a straight line with the number of rows.

Input Size (n)	Approx. Operations
10	About 30 (3 steps x 10 rows)
100	About 300 (3 steps x 100 rows)
1000	About 3000 (3 steps x 1000 rows)

Pattern observation: The total work grows linearly as the data size grows.

Final Time Complexity

Time Complexity: O(n)

This means the time needed grows in direct proportion to the number of rows in the data.

Common Mistake

[X] Wrong: "Combining multiple steps multiplies the time complexity, making it much slower than each step alone."

[OK] Correct: Each step runs one after another, so the total time adds up, not multiplies. The overall growth stays linear, just with a bigger constant factor.

Interview Connect

Understanding how multiple data cleaning steps add up helps you explain your approach clearly and shows you can think about efficiency in real projects.

Self-Check

"What if we combined the cleaning steps into one function that processes all columns at once? How would the time complexity change?"

Practice

1. What is the main benefit of combining multiple cleaning steps in pandas using method chaining?

easy

A. It creates multiple copies of the data.

B. It slows down the data processing.

C. It makes the code shorter and easier to read.

D. It hides errors in the data.

Solution

1. Step 1: Understand method chaining

   Method chaining lets you apply many cleaning steps one after another in one line or block.

2. Step 2: Benefits of chaining

   This approach reduces code length and improves readability by avoiding intermediate variables.

3. Final Answer:

   It makes the code shorter and easier to read. -> Option C

4. Quick Check:

   Method chaining = shorter, readable code [OK]

Hint: Chaining keeps code neat and easy to follow [OK]

Common Mistakes:

• Thinking chaining slows down processing
• Believing it creates multiple data copies
• Assuming it hides data errors

2. Which of the following is the correct syntax to chain pandas cleaning methods to drop missing values and then reset the index?

easy

A. df.dropna().reset_index(drop=True

B. df.dropna.reset_index(drop=True)

C. df.dropna().reset_index(drop=True))

D. df.dropna().reset_index(drop=True)

Solution

1. Step 1: Check method calls

   Each method must be called with parentheses: dropna() and reset_index().

2. Step 2: Verify parentheses balance

   df.dropna().reset_index(drop=True) has balanced parentheses and correct chaining syntax.

3. Final Answer:

   df.dropna().reset_index(drop=True) -> Option D

4. Quick Check:

   Method calls need parentheses [OK]

Hint: Always use () after pandas methods when chaining [OK]

Common Mistakes:

• Missing parentheses after methods
• Unbalanced parentheses causing syntax errors
• Using incorrect method names

3. Given the DataFrame df below, what will be the output after applying the chained cleaning steps?

import pandas as pd
df = pd.DataFrame({
  'A': [1, 2, None, 4],
  'B': ['x', None, 'y', 'z']
})
cleaned = df.dropna().reset_index(drop=True)

medium

A. {'A': [1.0, 4.0], 'B': ['x', 'y']}

B. {'A': [1.0, 4.0], 'B': ['x', 'z']}

C. {'A': [1.0, 2.0, 4.0], 'B': ['x', 'y', 'z']}

D. {'A': [1.0, 2.0, 4.0], 'B': ['x', None, 'z']}

Solution

1. Step 1: Apply dropna()

   Rows with any missing values are removed. Row 1 and 2 have missing values, so only rows 0 and 3 remain.

2. Step 2: Reset index

   Resetting index with drop=True gives new index 0 and 1 for the remaining rows.

3. Final Answer:

   {'A': [1.0, 4.0], 'B': ['x', 'z']} -> Option B

4. Quick Check:

   dropna removes rows with NaN, reset_index fixes index [OK]

Hint: dropna removes rows with NaN, reset_index fixes numbering [OK]

Common Mistakes:

• Forgetting dropna removes rows with any NaN
• Not resetting index after dropping rows
• Assuming missing values are filled automatically

4. The following code is intended to clean a DataFrame by filling missing values with 0 and then dropping duplicates. What is the error?

cleaned = df.fillna(0).drop_duplicates

medium

A. Missing parentheses after drop_duplicates method.

B. fillna should be called after drop_duplicates.

C. drop_duplicates does not exist in pandas.

D. fillna cannot be chained with drop_duplicates.

Solution

1. Step 1: Check method calls

   fillna(0) is called correctly with parentheses, but drop_duplicates is missing parentheses.

2. Step 2: Understand method chaining syntax

   All methods must be called with parentheses to execute them.

3. Final Answer:

   Missing parentheses after drop_duplicates method. -> Option A

4. Quick Check:

   Methods need () to run [OK]

Hint: Always add () after pandas methods when chaining [OK]

Common Mistakes:

• Forgetting parentheses after methods
• Changing method order unnecessarily
• Thinking drop_duplicates is not a pandas method

5. You have a DataFrame df with columns 'Name', 'Age', and 'Score'. Some 'Age' values are missing, and some rows are duplicates. You want to fill missing 'Age' with the average age, remove duplicates, and reset the index. Which chained code correctly does this?

hard

A. df.fillna({'Age': df['Age'].mean()}).drop_duplicates().reset_index(drop=True)

B. df.drop_duplicates().fillna(df['Age'].mean()).reset_index(drop=True)

C. df.fillna(df['Age'].mean()).drop_duplicates.reset_index(drop=True)

D. df.reset_index(drop=True).fillna({'Age': df['Age'].mean()}).drop_duplicates()

Solution

1. Step 1: Fill missing 'Age' with average

   Use fillna with a dictionary to fill only 'Age' column with its mean.

2. Step 2: Remove duplicates and reset index

   Then drop duplicates and reset index with drop=True to clean the DataFrame fully.

3. Final Answer:

   df.fillna({'Age': df['Age'].mean()}).drop_duplicates().reset_index(drop=True) -> Option A

4. Quick Check:

   Fill missing, drop duplicates, reset index in order [OK]

Hint: Fill missing with dict, then drop duplicates, reset index last [OK]

Common Mistakes:

• Filling all columns instead of only 'Age'
• Missing parentheses on drop_duplicates
• Resetting index before cleaning data