Databricks Machine Learning Associate Question 40
Select 3You are tasked with setting up a Databricks cluster to train a machine learning model using libraries like TensorFlow, PyTorch, and Scikit-learn. The cluster needs to be optimized for machine learning workloads. Which of the following steps should you take to ensure the cluster meets the requirements?
- A
Select a Databricks Runtime version that includes 'ML' in its name, such as 'Databricks Runtime for Machine Learning'.
- B
Manually install TensorFlow, PyTorch, and Scikit-learn libraries via the cluster libraries interface after creating the cluster.
- C
Ensure the cluster uses a driver and worker node configuration with sufficient memory and compute resources for training.
- D
Enable the GPU acceleration checkbox to utilize GPU resources for the machine learning workloads if the selected Databricks Runtime supports GPUs.
- E
Select the default Databricks Runtime version without 'ML' in its name and manually configure the environment for machine learning.
Show answer and explanation
Correct answers: A, C, D
Explanation
To create a cluster optimized for machine learning workloads, it is best to use the Databricks Runtime for Machine Learning, which includes pre-installed libraries and is tailored for such tasks. Additionally, ensuring sufficient compute and memory resources is important for model training. Enabling GPU acceleration, if supported, can further enhance performance for applicable workloads.
- A. Correct.
Correct: The Databricks Runtime for Machine Learning includes pre-installed machine learning libraries like TensorFlow, PyTorch, and Scikit-learn, which simplifies the setup process.
- B. Incorrect.
Incorrect: The Databricks Runtime for Machine Learning already includes key libraries like TensorFlow, PyTorch, and Scikit-learn, so manually installing them is unnecessary.
- C. Correct.
Correct: Proper memory and compute resources are essential for efficiently training machine learning models on Databricks.
- D. Correct.
Correct: GPU acceleration can significantly enhance performance for compatible machine learning workloads, provided the runtime supports GPUs.
- E. Incorrect.
Incorrect: While the default Databricks Runtime can technically be configured for machine learning, it requires significantly more manual effort and lacks the pre-installed optimizations of the Machine Learning runtime.