Q: 7
A company that has hundreds of data scientists is using Amazon SageMaker to create ML models.
The models are in model groups in the SageMaker Model Registry.
The data scientists are grouped into three categories: computer vision, natural language processing
(NLP), and speech recognition. An ML engineer needs to implement a solution to organize the
existing models into these groups to improve model discoverability at scale. The solution must not
affect the integrity of the model artifacts and their existing groupings.
Which solution will meet these requirements?
Options
Discussion
A for this. Tags give that extra layer of sorting without changing the model groups, so you keep existing structure. If you moved models to new groups (like B) you'd break the rule about not touching current groupings. Pretty sure that's what AWS expects here, but open to debate if I'm missing something.
It’s A, not B-moving models would mess with grouping integrity which the question says to avoid. Tags keep existing groupings untouched while adding category info. Pretty common exam trap here for thinking Model Groups are flexible.
A tbh, tags let you organize without messing with current groups or model data. Fits the "don't affect integrity" bit.
I don’t think it’s B. A is correct since adding custom tags won’t mess with current model groups or artifacts. Tags let you organize models without moving anything. Saw a similar pattern in practice questions.
Option B
A is the way to go here since tags let you organize models by category without having to mess with their existing groupings or move anything around. Model groups already exist, so adding custom tags (like category) just lets you slice and search the registry better. I think this lines up with how AWS recommends resource grouping. Anyone see a downside?
A tbh, tagging is the AWS way to organize stuff without messing with artifact integrity or moving models around. Model groups are already in use so tags just add another filter layer. Pretty sure no need to restructure here.
A is right here because adding custom tags lets you organize models for search and filtering without disrupting existing model group structures or touching the artifacts. AWS recommends tags for this sort of flexible, non-intrusive organization. I think that's what they want, but if anyone reads the question differently let me know.
A imo, since using tags lets you add that extra layer of categorization without messing with how models are already grouped. Moving models into new groups (like B) would change existing associations, which the question says to avoid. I think this is a standard AWS tagging use case but happy to hear other thoughts.
Had something like this in a mock, and pretty sure A is right for keeping model groups intact.
Be respectful. No spam.
