Answer-first summary for fast verification
Answer: Convert the images to TFRecords and store them in a Cloud Storage bucket. Read the TFRecords by using the tf.data.TFRecordDataset function.
The correct answer is C. Converting the images to TFRecords and storing them in a Cloud Storage bucket is the most efficient option for reading images at scale during training. The TFRecord format is optimized for TensorFlow, allowing for efficient data serialization and parallel reading. Using the tf.data.TFRecordDataset function will enable you to stream the data efficiently, minimizing I/O bottlenecks. While other options might work, they come with drawbacks such as additional network overhead or manual loading and parsing, making TFRecords the most scalable and performance-oriented solution.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
As an ML engineer at your company, you are tasked with developing a model to identify your company’s products in images. You have access to over one million high-resolution product images stored in a Cloud Storage bucket. Your goal is to experiment with multiple TensorFlow models using Vertex AI Training. To ensure efficient training and minimize data I/O bottlenecks, you need a strategy to read images at scale during the training process. What should you do?
A
Load the images directly into the Vertex AI compute nodes by using Cloud Storage FUSE. Read the images by using the tf.data.Dataset.from_tensor_slices function.
B
Create a Vertex AI managed dataset from your image data. Access the AIP_TRAINING_DATA_URI environment variable to read the images by using the tf.data.Dataset.list_files function.
C
Convert the images to TFRecords and store them in a Cloud Storage bucket. Read the TFRecords by using the tf.data.TFRecordDataset function.
D
Store the URLs of the images in a CSV file. Read the file by using the tf.data.experimental.CsvDataset function.
No comments yet.