Google Professional Machine Learning Engineer Question 37
Select 3Google Cloud PlatformYou are tasked with building a content moderation application for a social media platform. The application should automatically detect and flag inappropriate images and text in real-time. Your team decides to leverage Google Cloud’s Model Garden to accelerate development. What steps should you take to implement this solution using pre-trained ML APIs available in Model Garden?
- A
Use the Vision API from Model Garden to detect inappropriate content in images and integrate it into your application.
- B
Train a custom model on TensorFlow for text moderation and deploy it on Vertex AI.
- C
Use the Natural Language API from Model Garden to analyze text for inappropriate content and integrate it into your application.
- D
Download a pre-trained model from Model Garden and fine-tune it for your specific content moderation use case.
- E
Leverage the Model Garden APIs directly by calling pre-trained models for both image and text content moderation without additional training.
Show answer and explanation
Correct answers: A, C, E
Explanation
Leveraging pre-trained ML APIs from Model Garden is a practical and efficient approach for building a content moderation application. The Vision API can handle image moderation tasks, while the Natural Language API is suited for text analysis. Additionally, Model Garden allows developers to directly call pre-trained models via APIs, eliminating the need for complex model training or fine-tuning in this scenario.
- A. Correct.
Correct: The Vision API available in Model Garden can be used to detect inappropriate content in images, making it an efficient choice for this use case.
- B. Incorrect.
Incorrect: While training a custom model is an option, leveraging pre-trained APIs from Model Garden is faster and more cost-effective for this scenario.
- C. Correct.
Correct: The Natural Language API from Model Garden can analyze text for inappropriate content, which aligns with the requirements of this application.
- D. Incorrect.
Incorrect: While fine-tuning may be necessary in some cases, the pre-trained ML APIs from Model Garden are designed to be used out-of-the-box for common use cases, avoiding the need for fine-tuning.
- E. Correct.
Correct: Model Garden provides pre-trained models that can be directly called through APIs for tasks like image and text moderation, simplifying the integration process.