Practice Test 2 | Google Cloud Certified Professional Cloud Architect | Dumps | Mock Test
Helicopter Racing League (HRL) wants to expand its use of managed AI and ML services to facilitate race predictions. Currently, race predictions are performed using TensorFlow running on VMs in the current public cloud provider.
Which GCP Services could host HRL TensorFlow models in a fully managed way (pick 2)?
A. AutoML Vision Edge
B. DialogFlow
C. AI Platform (now Vertex AI)
D. Kubeflow
E. BigQuery ML
F. TensorFlow Enterprise
Correct Answers: C and E
Google Cloud has many services for training and hosting TensorFlow models in the cloud in a managed way:
- AI Platform (now Vertex AI), that can train, host and use your ML model to make predictions at scale. You may choose the image of the instances, fully managed by GCP. Vertex AI is an integrated suite of products that comprends AI Platform functions together with pre-trained, AutoML and custom tooling
- BigQuery ML can make predictions with imported TensorFlow models, in addition to many other kinds of models
- A is wrong because AutoML Vision Edge helps to deploy models run on local devices such as smartphones, IoT devices. In our case Cloud Services are required.
- B is wrong because DialogFlow is for Natural Language Dialogs, not Tensorflow applied to videos
- D is wrong because Kubeflow is an open source library and tools for machine learning (ML) workflow deployments on Kubernetes
- F is wrong because TensorFlow Enterprise provides a scalable and managed TensorFlow development environment, not a hosting service.
For any further detail:
Comments are closed, but trackbacks and pingbacks are open.