Google Professional Machine Learning Engineer
Get started today
Ultimate access to all questions.
Comments
Loading comments...
Ultimate access to all questions.
As the Director of Data Science at a leading company, you've observed your team facing challenges while trying to integrate their Python code with the Kubeflow Pipelines SDK. The team is looking for a solution that not only streamlines the integration process but also ensures scalability and cost-effectiveness. Additionally, the solution should leverage the existing infrastructure without requiring significant changes. Which of the following strategies would you recommend to achieve these objectives? (Choose two correct options)
A
Install the custom Python code to Cloud Functions, and use Kubeflow Pipelines to trigger the Cloud Function. This approach is cost-effective but may not fully leverage the Kubeflow Pipelines SDK's capabilities.
B
Employ the predefined components available in the Kubeflow Pipelines SDK to access Dataproc, and run the custom code there. This method is straightforward but may not be the most scalable or cost-effective for large-scale data processing tasks.
C
Use the func_to_container_op function to construct custom components from the Python code. This approach efficiently integrates Python code into the SDK by creating custom components, leveraging the SDK's orchestration capabilities.
D
Enclose the custom Python code into Docker containers, and apply the load_component_from_file function to import the containers into the pipeline. This method offers flexibility and scalability but requires additional Docker expertise.
E
Combine the use of func_to_container_op for simple integrations and Docker containers for more complex scenarios to achieve both simplicity and scalability. This hybrid approach leverages the strengths of both methods.