Question 1

In a Pandas DataFrame, what would be the most efficient way to group the data by one column and apply multiple aggregate functions?

Options :

A : df.groupby('column').agg(['sum', 'mean'])

B : df.groupby('column').apply(lambda x: x.sum())

df.groupby('column').apply(lambda x: x.mean())

C : df.groupby('column').sum()

df.groupby('column').mean()

D : df['column'].sum()

df['column'].mean()

Answer: A

Question 2

You're using a Decision Tree classifier to predict whether a transaction is fraudulent. After training, you find that the classifier has a high accuracy on the training set but performs poorly on the test set. Which of the following techniques is most suitable to improve the model's performance?

Options :

A : Use a different kernel function.

B : Prune the decision tree.

C : Apply grid search without cross-validation.

D : Change the evaluation metric to accuracy.

Answer: B

Question 3

In a given dataset, you have a categorical feature "State" that includes the 50 U.S. states. You decide to one-hot encode this feature. What potential issue should you be cautious about?

Options :

A : Introducing multicollinearity

B : Reducing the dimensionality of the dataset

C : Creating an imbalanced dataset

D : Inducing underfitting in the model

Answer: A

Question 4

You are pulling data from various databases using Python to compile a large dataset for analysis. How would you ensure the data’s accuracy and reliability?

Options :

A : Use the most recent data available in all databases.

B : Write the Python script in a way that it runs faster to save computational resources.

C : Compare a random subset of data with source records to validate accuracy.

D : Perform exploratory data analysis (EDA) only after data cleaning is done.

Answer: C

Question 5

You notice that the transformation phase of the ETL process is extremely slow. You suspect that the issue lies in the inefficient handling of a large DataFrame in Pandas. Which of the following should be your first step to troubleshoot?

Options :

A : Break the data into smaller chunks and process them sequentially.

B : Switch to a different data processing library like Dask or PySpark.

C : Upgrade the hardware to speed up the process.

D : Review and optimize the Pandas code for performance.

Answer: D

Confidently Practice Online with Free PCED-30-01 Exam Cram

Buying Options: