QuestionQ70

VerifiedScaling prototypes into ML models

You developed a Transformer model in TensorFlow to translate text. Your training data includes millions of documents in a Cloud Storage bucket. You plan to use distributed training to reduce the training time. You need to configure the training job while minimizing the effort needed to modify code and manage the cluster configuration. What should you do?

Explanation

tf.distribute.MirroredStrategy provides synchronous data-parallel training across the multiple GPUs attached to one machine. A single Vertex AI worker pool using an A2 GPU machine avoids multi-worker cluster configuration while requiring only minimal changes to make the TensorFlow model strategy-aware.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!