Google Professional Machine Learning Engineer
Comments
Loading comments...
Ultimate access to all questions.
You are tasked with retraining a pre-trained EfficientNet model daily on a dataset of 20,000 images for an image classification project. The project has strict budget constraints, requiring minimization of infrastructure costs without compromising the efficiency of the retraining process. Additionally, the solution must ensure data is stored securely and is easily accessible for the retraining process. Considering these requirements, which of the following platform components and configuration environments should you choose? (Choose one correct option)
A
A Google Kubernetes Engine cluster with a V100 GPU Node Pool and an NFS Server, which provides high performance but may incur higher costs due to the complexity and maintenance of the cluster.
B
A Deep Learning VM with 4 V100 GPUs and local storage, offering high computational power but potentially higher costs and less scalability for storage needs.
C
An AI Platform Training job using a custom scale tier with 4 V100 GPUs and Cloud Storage, combining powerful GPU resources with cost-effective and scalable storage solutions.
D
A Deep Learning VM with 4 V100 GPUs and Cloud Storage, providing a balance between computational power and storage scalability but may not be as cost-effective as other options for daily retraining.