AI-900 exam dumps

AI-900 practice question 134 of 286

Microsoft Azure AI Fundamentals. Free level, Microsoft. Free question with the correct answer and a full explanation.

AI-900 Question 134

Select 2

You are a data scientist tasked with building and training a machine learning model on Azure. Your solution requires a scalable environment for model training and a managed service to orchestrate the training process. Which Azure services should you use to meet these requirements?

  1. A

    Azure Machine Learning

  2. B

    Azure Databricks

  3. C

    Azure Kubernetes Service (AKS)

  4. D

    Azure Cognitive Services

  5. E

    Azure Synapse Analytics

Show answer and explanation

Correct answers: A, B

Explanation

Azure Machine Learning provides a comprehensive managed environment for orchestrating the entire machine learning lifecycle, from training to deployment. For scalable training, Azure Databricks offers a powerful platform that integrates with Azure Machine Learning to handle large datasets and distributed training. These two services together meet the requirements of scalability and managed orchestration for machine learning workflows.

  • A. Correct.

    Azure Machine Learning is a managed service that provides tools for building, training, and deploying machine learning models. It is ideal for orchestrating the training process.

  • B. Correct.

    Azure Databricks is a scalable data analytics and machine learning platform based on Apache Spark, suitable for data preparation and large-scale model training.

  • C. Incorrect.

    Azure Kubernetes Service (AKS) is primarily used for container orchestration and is not specific to machine learning model training or orchestration.

  • D. Incorrect.

    Azure Cognitive Services provides pre-built AI models for tasks like vision, speech, and language. It is not used for training custom machine learning models.

  • E. Incorrect.

    Azure Synapse Analytics is a data integration and analytics service focused on data warehousing and big data analytics, but it does not provide direct support for managing or training machine learning models.

Timed practice exam

Take a AI-900 practice test under exam conditions

60 questions in 30 minutes, drawn from this bank, with a score report and a per-question review when you finish.

Start timed exam