Similar to SageMakerEstimator Is there a way to set hyperParameters map to XGBoostSageMakerEstimator

[sajjap]

System Information

• Spark
• sagemaker-spark-sdk Version
• Spark Version -spark_2.2.0-1.1.0
• Algorithm (XGBoost)

Describe the problem

Similar to SageMakerEstimator Is there a way to set hyperParameters map to XGBoostSageMakerEstimator, I do not see that it is an option in API docs. If not will you be able to enable this feature?

[knakad]

Hi @sajjap! Thanks for reaching out! Unfortunately, there are no immediate plans to add this in the near future. That being said, we're always reprioritizing our backlog based on customer feedback and will keep this in mind in the future.

Closing this issue as a duplicate of https://github.com/aws/sagemaker-spark/issues/55 and noting the additional feature request.

[sajjap]

Hi @knakad, Thanks for a quick response. But this isn't related to auto model tuning. I was referring to something similar to the highlighted in below example from sagemaker-spark github page but with the XGBoostSageMakerEstimator

Instead of creating a KMeansSageMakerEstimator, you can create an equivalent SageMakerEstimator:

val estimator = new SageMakerEstimator( trainingImage = "382416733822.dkr.ecr.us-east-1.amazonaws.com/kmeans:1", modelImage = "382416733822.dkr.ecr.us-east-1.amazonaws.com/kmeans:1", requestRowSerializer = new ProtobufRequestRowSerializer(), responseRowDeserializer = new KMeansProtobufResponseRowDeserializer(), hyperParameters = Map("k" -> "10", "feature_dim" -> "784"), sagemakerRole = IAMRole(roleArn), trainingInstanceType = "ml.p2.xlarge", trainingInstanceCount = 1, endpointInstanceType = "ml.c4.xlarge", endpointInitialInstanceCount = 1, trainingSparkDataFormat = "sagemaker")

[knakad]

Oops! Thanks for the callout =)

Oooh, I follow. Thanks for the clarification! We'll keep this open and revisit the priority :)