Question 1

You have a DataFrame df that contains duplicate rows. Your task is to remove duplicates based on two columns: 'A' and 'B'. What is the correct line of code to remove such duplicates while keeping the first occurrence?

Options :

A : df.drop_duplicates(subset=['A', 'B'], keep=False)

B : df.drop_duplicates()

C : df.unique(subset=['A', 'B'])

D : df.drop_duplicates(subset=['A', 'B'], keep='first')

Answer: D

Question 2

Suppose you are working on a classification problem with an imbalanced dataset. Which of the following techniques would not be effective for addressing the issue?

Options :

A : Using accuracy as the evaluation metric

B : Using the Synthetic Minority Over-sampling Technique (SMOTE)

C : Using different evaluation metrics such as F1-score.

D : Under-sampling the majority class

Answer: A

Question 3

You have a DataFrame named df that has three columns: 'A', 'B', 'C'. The DataFrame has 100 rows. Your task is to create a DataFrame where each column 'A', 'B', 'C' is transformed into 3 columns each: 'A_min', 'A_max', 'A_mean', 'B_min', 'B_max', 'B_mean', 'C_min', 'C_max', 'C_mean'. Which of the following code snippets accomplishes this task?

Options :

A : new_df = df.groupby(['A', 'B', 'C']).agg(['min', 'max', 'mean'])

B : new_df = df.apply(['min', 'max', 'mean'])

C : new_df = df.pivot(columns=['A', 'B', 'C'], values=['min', 'max', 'mean'])

D : new_df = df.aggregate(

{

'A': ['min', 'max', 'mean'],

'B': ['min', 'max', 'mean'],

'C': ['min', 'max', 'mean'],

}

)

Answer: D

Question 4

You have a linear regression model with 5 features. After checking the VIF (Variance Inflation Factor) for multicollinearity, one of the features has a VIF of 12. What should you consider doing?

Options :

A : Increase the sample size to improve the model.

B : Remove the feature with the high VIF from the model.

C : Use L1 regularization to mitigate the effect of multicollinearity.

D : Ignore it, as VIF of 12 is acceptable.

Answer: B

Question 5

While evaluating a logistic regression model using a confusion matrix, you notice that the True Positive rate is significantly higher than the True Negative rate. What does this observation imply?

Options :

A : The model is better at identifying the positive class than the negative class

B : The model performs equally well for both classes

C : The model has a high level of bias

D : The model is overfitting

Answer: A

Higher Test Marks with Free Online PCED Exam Practice

Buying Options: