Creating MultiIndex DataFrames in Pandas - Practice Exercises

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of MultiIndex creation from tuples

What is the output of this code snippet that creates a MultiIndex DataFrame from tuples?

Pandas

import pandas as pd

index = pd.MultiIndex.from_tuples([('A', 1), ('A', 2), ('B', 1)])
df = pd.DataFrame({'Value': [10, 20, 30]}, index=index)
print(df)

A

       Value
A 1      10
  2      20
B 1      30

B

Value
A 1 10
  2 20
B 1 30

C

   Value
A 1    10
  2    20
B 1    30

D

Value
A 1 10
A 2 20
B 1 30

💡 Hint

Look at how pandas displays MultiIndex DataFrames with hierarchical row labels.

The MultiIndex shows two levels: first level with 'A' and 'B', second level with 1 and 2. The DataFrame prints with indentation for the second level.

❓ data_output

intermediate

Number of rows in MultiIndex DataFrame

Given this MultiIndex DataFrame, how many rows does it contain?

Pandas

import pandas as pd

arrays = [['X', 'X', 'Y', 'Y'], [10, 20, 10, 20]]
index = pd.MultiIndex.from_arrays(arrays, names=('Letter', 'Number'))
df = pd.DataFrame({'Score': [5, 15, 10, 20]}, index=index)
print(len(df))

A 2

B 0

C 4

D 1

💡 Hint

Count the number of unique index pairs created.

The MultiIndex has 4 pairs: ('X',10), ('X',20), ('Y',10), ('Y',20). So the DataFrame has 4 rows.

🔧 Debug

advanced

Identify the error in MultiIndex creation

What error does this code raise when trying to create a MultiIndex DataFrame?

Pandas

import pandas as pd

index = pd.MultiIndex.from_tuples([('A', 1), ('B',)])
df = pd.DataFrame({'Value': [10, 20]}, index=index)

A TypeError: Cannot create MultiIndex from tuples

B No error, DataFrame created successfully

C IndexError: tuple index out of range

D ValueError: Length of tuples must be same

💡 Hint

Check if all tuples have the same length for MultiIndex.

MultiIndex.from_tuples requires all tuples to have the same length. Here, ('A', 1) has length 2 but ('B',) has length 1, causing ValueError.

🚀 Application

advanced

Create MultiIndex DataFrame from product of lists

Which code snippet correctly creates a MultiIndex DataFrame with all combinations of ['Red', 'Blue'] and [1, 2] as index levels?

A

import pandas as pd
index = pd.MultiIndex.from_product([['Red', 'Blue'], [1, 2]], names=['Color', 'Number'])
df = pd.DataFrame({'Value': [10, 20, 30, 40]}, index=index)
print(df)

B

import pandas as pd
index = pd.MultiIndex.from_tuples([('Red', 1), ('Blue', 2)])
df = pd.DataFrame({'Value': [10, 20]}, index=index)
print(df)

C

import pandas as pd
index = pd.MultiIndex.from_arrays([['Red', 'Blue'], [1, 2]], names=['Color', 'Number'])
df = pd.DataFrame({'Value': [10, 20]}, index=index)
print(df)

D

import pandas as pd
index = pd.MultiIndex.from_product([['Red', 'Blue'], [1, 2, 3]], names=['Color', 'Number'])
df = pd.DataFrame({'Value': [10, 20, 30, 40]}, index=index)
print(df)

💡 Hint

Use from_product to get all combinations of the two lists.

from_product creates the cartesian product of the lists, producing 4 index pairs. The DataFrame has 4 rows matching these pairs.

🧠 Conceptual

expert

Understanding MultiIndex level names and access

Given a MultiIndex DataFrame with levels named 'City' and 'Year', which code correctly selects all rows where 'City' is 'Paris'?

Pandas

import pandas as pd

index = pd.MultiIndex.from_tuples([('Paris', 2020), ('Paris', 2021), ('London', 2020)], names=['City', 'Year'])
df = pd.DataFrame({'Population': [2_100_000, 2_150_000, 8_900_000]}, index=index)

A df.loc[('Paris',)]

B df.xs('Paris', level='City')

C df.loc['Paris']

D df.loc[df.index.get_level_values('City') == 'Paris']

💡 Hint

Use the cross-section method to select by level name.

df.xs('Paris', level='City') selects all rows where the 'City' level equals 'Paris'. df.loc['Paris'] works only if 'City' is the outermost index level, which it is here, but xs is more explicit and reliable.