Knowledge Check 7_ ML Models_ ISYS7730
.pdf
keyboard_arrow_up
School
Boston University *
*We aren’t endorsed by this school
Course
3
Subject
Industrial Engineering
Date
Apr 3, 2024
Type
Pages
6
Uploaded by BarristerVulture3667
2/11/24, 8:03 PM
Knowledge Check 7: ML Models: ISYS7730.01|.02 Spring 2024 Data Analytics 3: In Practice [Beebe]
https://bostoncollege.instructure.com/courses/1657214/quizzes/2429043
1/6
*
Some questions not yet graded
Due
Feb 11 at 11:59pm
Points
15
Questions
7
Time Limit
None
Allowed Attempts
5
Instructions
Attempt History
Attempt
Time
Score
KEPT
Attempt 2
9 minutes
9 out of 15 *
LATEST
Attempt 2
9 minutes
9 out of 15 *
Attempt 1
41 minutes
4.5 out of 15 *
Correct answers will be available on Feb 12 at 8am.
Score for this attempt: 9
out of 15 *
This knowledge check is to help you assess if you need to review anything or ask follow-up questions to the
TAs/professor.
The last questions will ask you to use ChatGPT to create a graph in python. Before starting the quiz:
Read this short article on how to create an effective prompt: Prompts to Try Asking ChatGPT
(https://bostoncollege.instructure.com/courses/1657214/files/73222560?wrap=1) (https://bostoncollege.instructure.com/courses/1657214/files/73222560/download?download_frd=1)
If you have not done so already, create a free account to use ChatGPT:
Go to https://openai.com
(https://openai.com) Click "Try ChatGPT" in the upper right corner of the page
Click to Sign Up and enter the required information
As an example, here's my conversation with ChatGPT to figure out how to get data from YouTube's public API into a
pandas dataframe: My ChatGPT Conversation.pdf
(https://bostoncollege.instructure.com/courses/1657214/files/73222534?wrap=1) (https://bostoncollege.instructure.com/courses/1657214/files/73222534/download?download_frd=1)
KNOWLEDGE CHECK 7: ML MODELS
KNOWLEDGE CHECK 7: ML MODELS
Class 8: Predictive Modeling (2/7)
Take the Quiz Again
2/11/24, 8:03 PM
Knowledge Check 7: ML Models: ISYS7730.01|.02 Spring 2024 Data Analytics 3: In Practice [Beebe]
https://bostoncollege.instructure.com/courses/1657214/quizzes/2429043
2/6
Submitted Feb 11 at 8:02pm
This attempt took 9 minutes.
2 / 2 pts
Question 1
Overfitting is bad because
The model does not perform well on the data set you trained it on. The model might perform poorly when applied on a different dataset (out-of-
sample).
The model gives large coefficient values. The model takes too long to run the complicated models. Overfitting is a problem because, while the model performs well on
the data set you trained it on, it does not perform similarly well on
other datasets.
3 / 3 pts
Question 2
You are running a classification model to predict which of three animals (cat, horse,
dog) are in each photo in a dataset. You get the following confusion matrix: Actual
Actual
Actual
Cat
Horse
Dog
Predicted
Cat
85
8
18
Predicted
Horse
5
62
37
Predicted
Dog
30
3
100
2/11/24, 8:03 PM
Knowledge Check 7: ML Models: ISYS7730.01|.02 Spring 2024 Data Analytics 3: In Practice [Beebe]
https://bostoncollege.instructure.com/courses/1657214/quizzes/2429043
3/6
To 2 decimal places, what is the accuracy of your model? (Enter your answer as a
decimal value, not a percentage. So if your answer was, for example, 50/60, enter
0.83 not 83.33%.)
0.71
Sum the correct entries divided by the total entries.
0 / 1 pts
Question 3
Incorrect
Incorrect
Which statement is true about the following command:
sklearn.model_selection.train_test_split( features, outcome,
test_size = 0.33)
If my dataset has 300 observations, this command will assign the first 200
observations to the train set and the rest of the observations to the test set.
The output to this function consists of two outputs: train and test subsets. If my dataset has 300 observations, this command will create a train set of
100 observations.
This command splits features and outcome into test and train features and
outcome
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
- Access to all documents
- Unlimited textbook solutions
- 24/7 expert homework help