MENU
How many samples are recommended to train a custom ML model that has one custom extra field?
Suggested Answer: C
Get Premium UiPath-ABAv1 Questions
as Interactive Web-Based Practice Test or PDF
Real Exam Format
Detailed Explanations
Instant Access
Get Your Practice Test
To train a custom Machine Learning (ML) model that includes one custom extra field, it is recommended to use between 20 to 50 document samples. This range provides a sufficient amount of data to ensure the model can learn effectively while accommodating the variability introduced by the additional custom field
22 comments · newest first
MA
Marget
2 days ago
I disagree, 50-90 sounds more reliable for accuracy.
CH
Chana
7 days ago
Wait, only 5-10 samples? That seems too low.
EN
Ena
12 days ago
Definitely leaning towards option C!
DE
Delsie
17 days ago
I’ve heard 20-50 samples is a good range.
GW
Gwenn
23 days ago
Not sure about this, feels like we need more data for accuracy.
RE
Reynalda
28 days ago
Definitely leaning towards B, 5-20 sounds reasonable.
BR
Brandee
1 month ago
Wait, only 5-10 samples? That seems too low!
BE
Bethanie
1 month ago
I think 50-90 is way too much for just one extra field.
BU
Bulah
1 month ago
I heard 20-50 is a good range for custom fields.
HE
Henriette
2 months ago
50-90 sounds excessive for just one extra field!
DE
Deeanna
2 months ago
Wait, are we sure about needing that many samples?
JA
Janine
2 months ago
5-20 seems way too low for a custom field.
EU
Eun
2 months ago
Definitely leaning towards option C!
DA
Dahlia
2 months ago
I’ve heard 20-50 is a good range.
LO
Lon
2 months ago
This is tricky! I feel like the more samples, the better, but I can't remember if 50 is too much for just one extra field. Maybe C is safe?
TA
Tawanna
3 months ago
I recall something about needing a minimum of 10 samples for basic training, but with an extra field, it could be higher. I might go with B.
AR
Arlene
3 months ago
I'm not entirely sure, but I feel like I practiced a question that suggested 50 samples would be more reliable. Maybe D is the right choice?
RE
Reita
3 months ago
I think I remember reading that for a custom field, you might need at least 20 samples, so I'm leaning towards option C.
NA
Natalya
3 months ago
I’m a bit confused; I thought 5-20 was enough for simpler models, but I guess it depends on the complexity of the field.
RO
Ronna
3 months ago
I feel like I've seen recommendations that vary widely, but 50-90 seems excessive for just one extra field.
FR
Freeman
4 months ago
I think I practiced a question similar to this, and it suggested at least 20 samples, so I might lean towards option C.
TI
Timothy
5 months ago
I remember reading that more samples generally lead to better model performance, but I'm not sure how many are really needed for just one extra field.