Train-test split in ML Python - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Why do we use a train-test split in machine learning?

Choose the best reason why splitting data into training and testing sets is important.

A To evaluate how well the model performs on unseen data.

B To increase the size of the training data for better learning.

C To reduce the number of features in the dataset.

D To speed up the training process by using less data.

❓ Predict Output

intermediate

Output of train-test split sizes

What will be the output sizes of training and testing sets after this code runs?

ML Python

from sklearn.model_selection import train_test_split
X = list(range(100))
y = [x * 2 for x in X]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=42)
print(len(X_train), len(X_test))

A 75 25

B 25 75

C 70 30

D 80 20

❓ Hyperparameter

advanced

Choosing the test_size parameter

Which test_size value is best if you want to maximize training data but still have a reliable test set?

A 0.05

B 0.2

C 0.5

D 0.8

🔧 Debug

advanced

Identify the error in this train-test split code

What error will this code raise?

ML Python

from sklearn.model_selection import train_test_split
X = [1, 2, 3]
y = [4, 5]
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.33)

A No error, code runs successfully

B TypeError: test_size must be an integer or float

C SyntaxError: invalid syntax

D ValueError: Found input variables with inconsistent numbers of samples

❓ Model Choice

expert

Best practice for train-test split with imbalanced classes

You have a classification dataset with very imbalanced classes. Which train-test split approach is best to keep class proportions consistent?

A Split data manually by slicing the dataset in order.

B Randomly split data without stratify to avoid bias.

C Use train_test_split with stratify parameter set to the target labels.

D Use only training data and skip testing to avoid imbalance issues.