Why handling missing data matters in Pandas - Challenge Your Understanding

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the output of this code with missing data?

Consider the following pandas DataFrame with missing values. What will be the output of the code below?

Pandas

import pandas as pd
import numpy as np
df = pd.DataFrame({'A': [1, 2, np.nan, 4], 'B': [5, np.nan, np.nan, 8]})
result = df['A'].mean()
print(round(result, 2))

A 2.25

B 2.33

C nan

D 3.0

💡 Hint

Remember that pandas mean() skips missing values by default.

The mean() function in pandas ignores NaN values by default. So it calculates mean of 1, 2, and 4 which is (1+2+4)/3 = 2.33.

❓ data_output

intermediate

How many rows remain after dropping missing data?

Given this DataFrame, how many rows remain after dropping all rows with any missing values?

Pandas

import pandas as pd
import numpy as np
df = pd.DataFrame({'X': [1, np.nan, 3, 4], 'Y': [np.nan, 2, 3, 4]})
df_clean = df.dropna()
print(len(df_clean))

A 1

B 2

C 3

D 4

💡 Hint

dropna() removes rows with any NaN values.

Only the last row has no missing values, so after dropna(), only 1 row remains.

🔧 Debug

advanced

What error does this code raise when filling missing data?

What error will this code produce when trying to fill missing values?

Pandas

import pandas as pd
df = pd.DataFrame({'A': [1, 2, None], 'B': ['x', 'y', None]})
df['B'] = df['B'].fillna(0)
print(df)

A TypeError

B KeyError

C ValueError

D No error, fills missing with 0

💡 Hint

fillna can fill missing values with any scalar, even if column is string type.

fillna replaces None with 0 without error, even in string columns.

🧠 Conceptual

advanced

Why is handling missing data important before modeling?

Which of the following best explains why handling missing data is important before building a machine learning model?

A Models ignore missing data automatically, so no need to handle it.

B Missing data always improves model accuracy.

C Missing data can cause models to crash or produce biased results.

D Handling missing data slows down model training without benefits.

💡 Hint

Think about how missing values affect calculations and predictions.

Missing data can cause errors or bias in models, so it must be handled properly.

🚀 Application

expert

Which option produces the correct imputed DataFrame?

Given this DataFrame with missing values, which code snippet correctly fills missing numeric values with the column mean and missing categorical values with the mode?

Pandas

import pandas as pd
import numpy as np
df = pd.DataFrame({'num': [1, 2, np.nan, 4], 'cat': ['a', np.nan, 'a', 'b']})

A

df['num'].fillna(df['num'].mean(), inplace=True)
df['cat'].fillna(df['cat'].mode()[0], inplace=True)
print(df)

B

df.fillna(df.mean(), inplace=True)
df.fillna(df.mode().iloc[0], inplace=True)
print(df)

C

df['num'] = df['num'].fillna(df['num'].median())
df['cat'] = df['cat'].fillna(df['cat'].mode()[0])
print(df)

D

df.fillna(method='ffill', inplace=True)
print(df)

💡 Hint

Use mean for numeric and mode for categorical separately.

Option A correctly fills numeric NaNs with mean and categorical NaNs with mode. Others either misuse fillna or use wrong methods.