MENU
Refer to the exhibit:
The plots represent two models, A and B, being fit to the same two data sets, training and validation.
Model A is 90.5% accurate at distinguishing blue from red on the training data and 75.5% accurate at doing the same on validation dat
a. Model B is 83% accurate at distinguishing blue from red on the training data and 78.3% accurate at doing the same on the validation data.
Which of the two models should be selected and why?
Suggested Answer: C
Get Premium A00-240 Questions
as Interactive Web-Based Practice Test or PDF
Real Exam Format
Detailed Explanations
Instant Access
Get Your Practice Test
30 comments · newest first
LA
Laura
11 months ago
Not sure if complexity is worth it if validation drops like that.
ST
Starr
11 months ago
Model A is definitely the way to go, higher overall accuracy!
KA
Kayleigh
11 months ago
Surprised that Model A is so much lower on validation.
LA
Lavelle
12 months ago
I think Model B is better for validation accuracy!
JA
Janey
12 months ago
Model A has better training accuracy, but validation is lower.
CE
Celestine
1 year ago
I thought we were supposed to prioritize validation performance, but Model A's training accuracy is really tempting. I'm confused about which one to pick!
GR
Graciela
1 year ago
I feel like Model B might be the better choice since it has a higher validation accuracy, even though its training accuracy is lower.
LE
Leonard
1 year ago
I remember a similar question where we had to pick a model based on overfitting. Model A seems to be overfitting since it has a much higher training accuracy.
KE
Kenda
1 year ago
I'm not entirely sure, but I think we should consider the validation accuracy more than the training accuracy when choosing a model.
PA
Pamella
1 year ago
I'm leaning towards Model B as well. The lower training accuracy but higher validation accuracy indicates it's less prone to overfitting.
LO
Lorita
1 year ago
Based on the information provided, I would select Model B. It has a higher validation accuracy, which suggests it will generalize better to new data.
JO
Jose
1 year ago
I'm a bit confused about the difference between training and validation accuracy. Do I need to consider both or just focus on the validation results?
MA
Mattie
1 year ago
Okay, so Model A has higher training accuracy but lower validation accuracy compared to Model B. I think I need to consider the concept of overfitting here.
DE
Delmy
1 year ago
Hmm, this is a tricky one. I'll need to carefully compare the training and validation accuracy for each model to decide which one is better.
ER
Erinn
1 year ago
I'd have to go with Model B. Sure, it's not as flashy as Model A, but it's got the chops where it counts - on the validation data. Accuracy on the training set is cool and all, but real-world performance is what matters in the end.
LA
Layla
1 year ago
Model B is simpler and still performs better on the validation data, that's what matters most.
TO
Torie
1 year ago
Yeah, Model A may be more accurate on the training data, but Model B is more accurate on the validation data.
EL
Elvis
1 year ago
I agree, Model B seems like the better choice for real-world performance.
DE
Desiree
1 year ago
Haha, this is a classic case of the 'validation gap'! Model A is clearly the diva of the group, strutting its stuff on the training data but falling flat on the validation. Model B is the humble workhorse, getting the job done without all the drama.
TY
Tyisha
2 years ago
Hmm, I'm not convinced. Model B may be simpler, but it's still got a pretty decent accuracy on the validation data. I'd go with the simpler model, it's less likely to have issues with overfitting.
JO
Joaquin
1 year ago
But accuracy is also important, especially on validation data.
QU
Quentin
1 year ago
I agree, simplicity is key when it comes to model selection.
NO
Noah
2 years ago
I think Model A is the better choice. Sure, it has a higher accuracy on the training data, but that's probably because it's more complex and able to overfit the training set. The validation data is a better representation of real-world performance, and Model A does better there.
VE
Veronika
1 year ago
Model B. It is simpler with a higher accuracy than model A on validation data.
LE
Leota
1 year ago
I agree. Model A is likely overfitting the training data, so Model B is a better choice for generalization.
SU
Susana
1 year ago
Model A. It is more complex with a higher accuracy than model B on training data.
RO
Rosalind
1 year ago
Model A. It performs better on the boundary for the training data.
JE
Jesusa
2 years ago
That's true, but Model A is more complex and performs better on the training data boundary.
NI
Nieves
2 years ago
But Model B has higher accuracy on the validation data, so maybe we should consider that.
JE
Jesusa
2 years ago
I think we should select Model A because it has higher accuracy on the training data.