Exam AWS Certified Machine Learning - Specialty topic 1 question 113 discussion

I would select B. Based on the following AWS documentation it appears this is the right approach: https://sagemaker.readthedocs.io/en/stable/frameworks/tensorflow/using_tf.html https://github.com/aws-samples/amazon-sagemaker-script-mode/blob/master/tf-horovod-inference-pipeline/train.py

B is my answer. Reading Data filenames = ["s3://bucketname/path/to/file1.tfrecord", "s3://bucketname/path/to/file2.tfrecord"] dataset = tf.data.TFRecordDataset(filenames)

This approach leverages the existing TFRecord format and minimizes changes to the current setup, ensuring a smooth transition to using Amazon SageMaker with minimal development effort.

option B

Letters C and D need code development and are therefore discarded. As we want a scalable data storage, it is recommended to use the Letter B, since S3 is scalable. Letter A is wrong as your personal computer is not scalable.

Where had the Capslock Donald gone? I kinda miss his answers

Internet connectivity issue: then how IOT can be a solution? (Correct answer should be A)

Amazon SageMaker script mode enables training a machine learning model using a script that you provide. By using the unchanged train.py script and putting the TFRecord data into an Amazon S3 bucket, you can easily point the Amazon SageMaker training invocation to the S3 bucket without reformatting the training data. This option avoids the need to rewrite the train.py script or to prepare the data in a different format. It also leverages the scalability and cost-effectiveness of Amazon S3 for storing large amounts of data, which is important for training machine learning models.

B, obviously

I like answer B

Why not A? Why can't we train it from local?

B. https://aws.amazon.com/about-aws/whats-new/2019/01/amazon-sagemaker-batch-transform-now-supports-tfrecord-format/

Unfortunilty you can't use the script unchanged, there are some things that need to be added: 1. Make sure your script can handle --model_dir as an additional command line argument. If you did not specify a location when you created the TensorFlow estimator, an S3 location under the default training job bucket is used. Distributed training with parameter servers requires you to use the tf.estimator.train_and_evaluate API and to provide an S3 location as the model directory during training. 2. Load input data from the input channels. The input channels are defined when fit is called. ## https://sagemaker.readthedocs.io/en/stable/frameworks/tensorflow/using_tf.html Beause of the pre-rec Ans A and B are an easy disqualifcation. There is no need to change the training format so option C is a red herring Ans is D Not the most obvious answer