Answer-first summary for fast verification
Answer: Create an experiment in Kubeflow Pipelines to organize multiple runs.
The best approach is to create an experiment in Kubeflow Pipelines to organize multiple runs. Kubeflow Pipelines provides a structured and scalable way to manage and compare multiple training runs with different model architectures and hyperparameters. It allows you to track progress and compare results efficiently. Option A (AutoML Tables) does not provide the flexibility for custom model architectures. Option B (Cloud Composer) is more suitable for orchestrating workflows but lacks detailed tracking and comparison features for multiple runs. Option C (AI Platform with similar job names) would not be effective for organizing and comparing the results systematically. Therefore, the correct answer is D.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
You are tasked with designing a customized deep neural network using Keras to predict customer purchases based on their purchase history. Your goal is to explore the model performance by using multiple model architectures, store the training data, and be able to compare the evaluation metrics in a unified dashboard. Considering that you need a scalable solution that can organize and track multiple runs effectively, what should you do?
A
Create multiple models using AutoML Tables.
B
Automate multiple training runs using Cloud Composer.
C
Run multiple training jobs on AI Platform with similar job names.
D
Create an experiment in Kubeflow Pipelines to organize multiple runs.