MENU
A company has built an image classification model to predict plant diseases from photos of plant leaves. The company wants to evaluate how many images the model classified correctly.
Which evaluation metric should the company use to measure the model's performance?
Suggested Answer: B
Get Premium AIF-C01 Questions
as Interactive Web-Based Practice Test or PDF
Real Exam Format
Detailed Explanations
Instant Access
Get Your Practice Test
30 comments · newest first
RO
Rosendo
11 months ago
Learning rate? That doesn't even measure performance!
LE
Lemuel
11 months ago
RMSE is for regression, not classification!
RY
Ryann
11 months ago
Wait, why not use R-squared? Isn't that useful too?
DE
Desirae
11 months ago
Definitely, accuracy makes sense for classification tasks.
LO
Lorita
12 months ago
I think accuracy is the way to go for this!
MY
Myong
12 months ago
Learning rate doesn't seem like an evaluation metric at all; it's more about how the model learns during training, right?
RO
Ronna
1 year ago
I feel like RMSE is also not suitable here, but I can't recall why exactly. It seems more for continuous values.
AL
Albert
1 year ago
I'm not entirely sure, but I remember we discussed R-squared being more relevant for regression tasks, not classification.
PA
Pauline
1 year ago
I think the accuracy metric might be the right choice since we're looking at how many images were classified correctly.
CH
Charisse
1 year ago
I'm a bit unsure about this one. I know accuracy is a common metric for classification models, but I'm wondering if there might be some nuance I'm missing. Maybe I should review the definitions of the other options just to be sure.
AN
Annamae
1 year ago
Okay, let me think this through. Accuracy seems like the most logical choice since we're evaluating a classification model. R-squared is more for regression problems, and RMSE is also for regression. Learning rate is a hyperparameter, not an evaluation metric. I'm pretty confident accuracy is the way to go here.
ZA
Zack
1 year ago
Accuracy makes sense, but I'm not sure if that's the only option. I might need to think through the other choices to see if any of them could also work for this type of model.
MA
Margurite
1 year ago
Hmm, this seems pretty straightforward. I think accuracy would be the best metric to use here since the goal is to measure how many images the model classified correctly.
RY
Ryann
2 years ago
D) Learning rate? Haha, that's clearly not the right metric for evaluating image classification. Might as well go with a random number generator for that one.
GE
Gerald
2 years ago
C) Root mean squared error (RMSE)
CO
Corinne
2 years ago
B) Accuracy
DO
Douglass
2 years ago
A) R-squared score
JE
Jennifer
2 years ago
I'm torn between B) Accuracy and C) RMSE. Guess I'll have to do some more research to decide.
TA
Tamala
2 years ago
RMSE is more suitable for regression tasks, so I would stick with Accuracy for this case.
OL
Olive
2 years ago
Accuracy is definitely the way to go for evaluating the performance of your image classification model.
MA
Mabel
2 years ago
I agree, Accuracy is a good choice for evaluating image classification models.
LA
Latosha
2 years ago
I think you should go with B) Accuracy. It's a common metric for classification models.
NI
Nilsa
2 years ago
I'd say C) Root mean squared error (RMSE) is the way to go. It gives a more nuanced view of the model's performance.
BU
Buck
2 years ago
Definitely going with B) Accuracy. It's the most straightforward way to measure how many images the model correctly classified.
NA
Nathan
2 years ago
User 4: Accuracy it is then, let's see how well the model performed.
AL
Alita
2 years ago
User 3: I agree, it's the most straightforward metric to use.
IN
In
2 years ago
User 2: Yeah, accuracy is the best way to measure performance.
MI
Micaela
2 years ago
User 1: I think we should go with B) Accuracy.
DE
Destiny
2 years ago
I agree with Dulce. Accuracy would be the best metric to measure how many images the model classified correctly.
DU
Dulce
2 years ago
I think the company should use Accuracy as the evaluation metric.