MENU
When should the model be retrained in the ML pipeline?
Suggested Answer: B
Get Premium AIP-210 Questions
as Interactive Web-Based Practice Test or PDF
Real Exam Format
Detailed Explanations
Instant Access
Get Your Practice Test
When concept drift is detected in the pipeline, it means that the model performance has degraded over time due to changes in the underlying data generating process. This requires retraining the model with new data that reflects the current situation and updating the model parameters accordingly. Reference:Use pipeline parameters to retrain models in the designer - Azure Machine Learning | Microsoft Learn,Retraining Model During Deployment: Continuous Training and Continuous Testing
23 comments · newest first
CR
Craig
4 days ago
I’d prioritize B and C. They directly impact performance.
DE
Denny
9 days ago
D seems less critical. Outliers can be managed without retraining.
SI
Simona
14 days ago
True, but if the model is drifting, B is more urgent.
LA
Lashaunda
19 days ago
C makes sense too. More data can improve accuracy.
SH
Shantay
25 days ago
Agreed! B is definitely a top reason to retrain.
KI
Kip
30 days ago
I think B is crucial. Concept drift means the model is outdated.
LY
Lynda
1 month ago
C is good, but I’d prioritize B over it.
TR
Trevor
1 month ago
I lean towards B too. It directly impacts performance.
RE
Rene
2 months ago
D could be a sign, but it’s not always necessary to retrain.
LO
Louann
2 months ago
A new monitoring component? That’s not a strong reason to retrain.
KR
Kristofer
2 months ago
True, but if the model is drifting, it needs retraining first.
MA
Marjory
2 months ago
C is also important. More data can improve accuracy.
QU
Quiana
2 months ago
Agreed! B is definitely the top choice.
DO
Dorethea
2 months ago
I think B is crucial. Concept drift means the model is outdated.
SC
Scot
3 months ago
I agree with B, but C is a close second for me!
NI
Nieves
3 months ago
Wait, outliers in live data? Shouldn't we just handle those instead?
CA
Casie
3 months ago
A new monitoring component? Not sure that warrants retraining.
FL
Florinda
3 months ago
I think C is also important. More data can improve accuracy.
TR
Tran
3 months ago
Definitely B! Concept drift means the model needs an update.
LA
Latia
3 months ago
Adding a new monitoring component seems like it wouldn’t necessarily require retraining, so I’m leaning towards B or C.
PH
Phung
4 months ago
I feel like we had a practice question about outliers, but I can't recall if that means we should retrain.
HY
Hyun
4 months ago
I’m not entirely sure, but I think retraining is also important when new data comes in, like option C.
JA
Janine
4 months ago
I remember we discussed concept drift in class, so I think option B might be the right answer.