Your data science team is training a PyTorch model for image classification based on a pre- trained RestNet model. You need to perform hyperparameter tuning to optimize for several parameters. What sh

Question

Options

• AConvert the model to a Keras model, and run a Keras Tuner job.
• BRun a hyperparameter tuning job on AI Platform using custom containers.
• CCreate a Kuberflow Pipelines instance, and run a hyperparameter tuning job on Katib.
• DConvert the model to a TensorFlow model, and run a hyperparameter tuning job on AI Platform.

• A
  5% (3)
• B
  75% (42)
• C
  16% (9)
• D
  4% (2)

Explanation

The model is already built in PyTorch, so converting it to Keras (A) or TensorFlow (D) would require significant rework and could introduce errors or performance regressions. AI Platform (now Vertex AI Training) supports custom containers, meaning you can run hyperparameter tuning jobs with your native PyTorch code inside a Docker container. This preserves the original framework, ResNet architecture, and avoids conversion overhead. Kubeflow Pipelines with Katib (C) is a valid but more complex infrastructure setup that requires managing a Kubernetes cluster, making it a heavier solution than simply using the managed AI Platform custom container support.

No community discussion yet for this question.