wayground logo

Free Printable Worksheets

Font size

S
M
L
XL
Worksheets

Lesson 3 (Model Training)

Total questions: 20

Worksheet time: 10mins

Name
Class
Date
1.

Data wrangling is a one-time process

a)

True

b)

False

2.

Ways to handle missing data

a)

Drop missing columns

b)

Replace with zero

c)

Replace with mean

d)

All of the above

3.

Select the activity that is not a part of data wrangling

a)

Restructuring

b)

Grouping

c)

Model Training

d)

Normalization

4.

________ helps to package data into consumable formats

a)

Datastore

b)

Datasets

5.

Datasets create an in-memeory copy of the underlying data

a)

False

b)

True

6.

Select the linear dimensionality reduction technique based on linear mathematical calculations

a)

T-SNE

b)

PCA

c)

Feature embedding

d)

All of the above

7.

________ provides alert about data drift

a)

Datastore

b)

Dataset

c)

Data monitor

8.

Hyperparameters are not learnt from the data

a)

True

b)

False

9.

Select the hyperparameters

a)

Number of training epochs

b)

Weights/coefficients

c)

Number of hidden layers

d)

Learning rate

10.

Select the evaluation metric not used for classification

a)

RMSE

b)

Precision

c)

F1 Score

d)

Squared error

11.

Algorithm which outputs real continuous outputs

a)

Regression

b)

Classification

12.

Evaluation metrics for regression

a)

Mean Squared Error(MSE)

b)

Accuracy

c)

Mean-Absolute-Error (MAE).

d)

All of the above

13.

The phenomenon in which an ML algorithm is not capable of coping with very large numbers of features

a)

Features

b)

Curse of dimensionality

c)

Feature engineering

d)

Feature selection

14.

Potential benefits of feature engineering

a)

Faster model training time

b)

Smaller trained model size

c)

More appropriate features for some algorithms

d)

Improved model accuracy

15.

Deep learning depends on feature engineering much more than classical machine learning

a)

True

b)

False

16.

Example of covariate shift

a)

A change in customer behavior over time.

b)

Two features that used to be correlated are no longer correlated.

c)

Upstream process change

d)

A sensor breaks and starts providing inaccurate readings.

17.

Validation data is

a)

Data used to learn the values of the parameters.

b)

Data used to tune the values of the hyperparameters.

c)

Data used to check the performance of the final, fully trained model.

18.

Classify music as belonging to multiple groups refers to

a)

Multi-class multi-label classification

b)

Multi-class single-label classification

19.

In confusion matrices,the sum FP and TP tells us the number of actual positive cases in dataset

a)

True

b)

False

20.

Which of the following helps reduce bias for models?

a)

Stacking

b)

Boosting

c)

Bagging