Google Professional Machine Learning Engineer
Get started today
Ultimate access to all questions.
Your team is working on creating a machine learning (ML) solution in Google Cloud to classify support requests for one of your company's platforms. After analyzing the project requirements, you have chosen TensorFlow to build the classifier because it allows for full control over the model's code, serving, and deployment. To manage the ML lifecycle, you will be using Kubeflow pipelines. Additionally, to save time, you prefer to leverage existing resources and managed services rather than developing a completely new model from scratch. How should you proceed with building the classifier?
Your team is working on creating a machine learning (ML) solution in Google Cloud to classify support requests for one of your company's platforms. After analyzing the project requirements, you have chosen TensorFlow to build the classifier because it allows for full control over the model's code, serving, and deployment. To manage the ML lifecycle, you will be using Kubeflow pipelines. Additionally, to save time, you prefer to leverage existing resources and managed services rather than developing a completely new model from scratch. How should you proceed with building the classifier?
Explanation:
Option C is the correct answer. The key reason is that the requirements specify the use of TensorFlow, having full control over the model’s code, and leveraging existing resources to save time. Full control over the model code excludes options A and B, as they use managed services with less customizability. While option D uses an established text classification model, it does not incorporate transfer learning, which is crucial for adapting the model to your specific support request classification task. Therefore, using an established text classification model on AI Platform to perform transfer learning (option C) aligns best with the project requirements and constraints.