MENU
In order to perform honest assessment on a predictive model, what is an acceptable division between training, validation, and testing data?
Suggested Answer: D
Get Premium A00-240 Questions
as Interactive Web-Based Practice Test or PDF
Real Exam Format
Detailed Explanations
Instant Access
Get Your Practice Test
12 comments · newest first
TE
Tegan
12 months ago
50% training and 50% validation? That doesn't leave room for testing!
GE
Geraldine
12 months ago
Wait, can you really have 100% training? Sounds off.
DO
Dorthy
1 year ago
Definitely not A or B, those are just wrong.
RE
Renato
1 year ago
I think 60-20-20 is better for validation!
CA
Carlota
1 year ago
Typically, a 70-20-10 split is common.
LA
Lavina
1 year ago
Option D seems off to me; I thought you needed some testing data to evaluate the model properly.
MA
Ma
1 year ago
I feel like I've seen questions where the training set is larger, but I can't recall the exact percentages.
RO
Rory
1 year ago
I think having no validation data at all, like in option B, is definitely not a good practice.
LE
Lera
1 year ago
I remember that a common split is 70% training, 15% validation, and 15% testing, but I'm not sure if that's what they want here.
NI
Nickolas
1 year ago
I'm a bit confused on the differences between clearing, purging, and destruction. I'll need to review those concepts again before deciding on the most effective one.
MA
Marlon
1 year ago
Hmm, I'm not too sure about this one. I'll have to think it through carefully to decide between the options.
RE
Rebbeca
1 year ago
I'm leaning towards picking HTTPS too, citing its importance in web security, but I'm a bit confused about its relevance for video specifically.