Loading datasets (CSV, built-in datasets) in ML Python - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of loading a CSV with pandas

What is the output of the following code snippet when loading a CSV file with pandas and printing the first 3 rows?

ML Python

import pandas as pd
from io import StringIO
csv_data = '''name,age,score
Alice,30,85
Bob,25,90
Charlie,35,88
David,40,92'''
df = pd.read_csv(StringIO(csv_data))
print(df.head(3))

A

    name  age  score
0   Alice   30     85
1     Bob   25     90
2  Charlie   35     88

B

88  53  eilrahC  2
09  52  boB  1
58  03  ecilA  0
erocs  ega  eman

C

   name  age  score
0  Alice  30  85
1  Bob  25  90
2  Charlie  35  88

D

   name  age  score
0  Alice  30  85
1  Bob  25  90
2  Charlie  35  88
3  David  40  92

❓ Model Choice

intermediate

Choosing dataset loading method for built-in datasets

You want to load the Iris dataset for a classification task using scikit-learn. Which code snippet correctly loads the dataset as a dictionary with data and target arrays?

A

import pandas as pd
data = pd.read_csv('iris.csv')
X, y = data.iloc[:, :-1], data.iloc[:, -1]

B

from sklearn.datasets import load_iris
data = load_iris()
X, y = data.data, data.target

C

from sklearn.datasets import fetch_iris
data = fetch_iris()
X, y = data.data, data.target

D

from sklearn.datasets import load_iris
data = load_iris(as_frame=True)
X, y = data.data, data.target

❓ Metrics

advanced

Evaluating dataset loading time

You load a large CSV dataset using pandas with and without specifying dtypes. Which statement about loading time and memory usage is true?

A Specifying dtypes reduces loading time but increases memory usage.

B Specifying dtypes increases loading time but reduces memory usage.

C Specifying dtypes has no effect on loading time or memory usage.

D Specifying dtypes reduces loading time and memory usage because pandas can optimize data storage.

🔧 Debug

advanced

Identifying error when loading CSV with missing header

What error will the following code raise when loading a CSV file without a header row using pandas without specifying header=null?

ML Python

import pandas as pd
from io import StringIO
csv_data = '''Alice,30,85
Bob,25,90
Charlie,35,88'''
df = pd.read_csv(StringIO(csv_data))
print(df.head())

A No error; pandas treats first row as header and data is loaded with column names 'Alice', '30', '85'

B TypeError because data types cannot be inferred

C ValueError because of mismatched columns

D ParserError due to missing header row

🧠 Conceptual

expert

Understanding dataset loading with train_test_split

You load a built-in dataset and split it into training and testing sets using scikit-learn's train_test_split. Which statement is true about the resulting datasets?

A train_test_split randomly shuffles data before splitting, so training and testing sets are representative samples.

B train_test_split splits data sequentially without shuffling, so testing set contains last samples only.

C train_test_split always splits data into equal halves regardless of test_size parameter.

D train_test_split requires the dataset to be pre-shuffled manually before splitting.