Memory usage analysis in Data Analysis Python - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Memory usage of pandas DataFrame columns

What is the output of the following code showing memory usage of each column in a pandas DataFrame?

Data Analysis Python

import pandas as pd
import numpy as np
df = pd.DataFrame({
    'A': np.arange(1000),
    'B': np.random.rand(1000),
    'C': ['text']*1000
})
mem_usage = df.memory_usage(deep=True)
print(mem_usage)

A

Index    128
A       1000
B       8000
C      40000
dtype: int64

B

Index    128
A       8000
B       8000
C      32000
dtype: int64

C

Index    128
A       8000
B       8000
C      61000
dtype: int64

D

Index    128
A       8000
B       1000
C      40000
dtype: int64

💡 Hint

Remember that integers and floats take 8 bytes each, and strings with repeated values can have shared memory but deep=True counts actual string memory.

The integer column 'A' has 1000 values * 8 bytes = 8000 bytes. The float column 'B' also has 8000 bytes. The string column 'C' has 1000 identical strings, each taking about 61 bytes including overhead, totaling 61000 bytes. The index memory is 128 bytes.

❓ data_output

intermediate

Memory usage reduction by changing data types

Given a DataFrame with 1 million integers stored as int64, what is the memory usage after converting the column to int8?

Data Analysis Python

import pandas as pd
import numpy as np
df = pd.DataFrame({'numbers': np.arange(1000000)})
mem_before = df.memory_usage(deep=True).numbers

df['numbers'] = df['numbers'].astype('int8')
mem_after = df.memory_usage(deep=True).numbers
print(mem_before, mem_after)

A 8000000 8000000

B 8000000 1000000

C 1000000 8000000

D 1000000 1000000

💡 Hint

int64 uses 8 bytes per value, int8 uses 1 byte per value.

Original memory is 1,000,000 values * 8 bytes = 8,000,000 bytes. After converting to int8, memory is 1,000,000 * 1 byte = 1,000,000 bytes.

🔧 Debug

advanced

Identify the cause of high memory usage in a DataFrame

Why does this DataFrame use more memory than expected?

import pandas as pd
df = pd.DataFrame({
    'col': ['apple', 'banana', 'apple', 'banana'] * 250000
})
print(df.memory_usage(deep=True))

A Because the DataFrame uses int64 for strings which is inefficient.

B Because the DataFrame index is very large and uses most memory.

C Because the DataFrame has many missing values increasing memory.

D Because the strings are stored as object dtype without using categorical type, so each string is stored separately.

💡 Hint

Check how pandas stores repeated strings in object dtype.

Object dtype stores each string separately, even if repeated. Using categorical dtype would reduce memory by storing unique strings once.

🚀 Application

advanced

Visualizing memory usage of DataFrame columns

Which code snippet correctly creates a bar plot showing memory usage of each column in a DataFrame df?

A

import matplotlib.pyplot as plt
mem = df.memory_usage(deep=True)
mem.plot(kind='bar')
plt.show()

B

import matplotlib.pyplot as plt
mem = df.memory_usage(deep=True)
plt.plot(mem)
plt.show()

C

import matplotlib.pyplot as plt
mem = df.memory_usage()
plt.bar(mem.index, mem.values)
plt.show()

D

import matplotlib.pyplot as plt
mem = df.memory_usage(deep=True)
plt.scatter(mem.index, mem.values)
plt.show()

💡 Hint

Use pandas plotting with kind='bar' for bar charts.

Option A uses pandas Series plot with kind='bar' which correctly shows a bar chart of memory usage per column. Option A misses deep=True and uses plt.bar directly but may not label axes well. Option A uses line plot which is less clear. Option A uses scatter plot which is not suitable here.

🧠 Conceptual

expert

Understanding memory usage impact of categorical data

Which statement best explains why converting a string column with many repeated values to categorical dtype reduces memory usage?

A Categorical dtype stores the unique strings once and replaces column values with integer codes, reducing memory.

B Categorical dtype compresses strings using zip compression internally, saving memory.

C Categorical dtype converts strings to fixed-length byte arrays, which always use less memory.

D Categorical dtype stores strings as pointers to the original strings, increasing memory usage.

💡 Hint

Think about how categorical data stores repeated values.

Categorical dtype stores unique values once and replaces each value with a small integer code, which uses less memory than storing full strings repeatedly.