[New Resource]: aws_sagemaker_monitoring_schedule

[wrbooth]

Description

We would like to be able to manage Sagemaker monitoring schedules through terraform. Some more detail on this capability can be found here:

https://docs.aws.amazon.com/sagemaker/latest/dg/model-monitor-scheduling.html

Requested Resource(s) and/or Data Source(s)

• aws_sagemaker_monitoring_schedule

Potential Terraform Configuration

resource "aws_sagemaker_monitoring_schedule" "my_schedule" {
  config = {
    monitoring_job_definition = {
      monitoring_app_specification = {
        image_uri = "ecr.awsamazon.com/my_image"
        container_arguments = ["--foo", "--bar"]
        container_entrypoint = ["/monitor/start", "--run"]
        post_analytics_processor_source_uri = "http://myanalyticsuri.com"
        record_preprocessor_source_uri = "http://mypreprocessoruri.com"
      }
      monitoring_inputs = [
        {
          endpoint_name = "my-inference-endpoint"
          local_path = "/mnt/endpoint"
          end_time_offset = "-PT7H"
          features_attribute = "feature"
          inference_attribute = "inference"
          probability_attribute = "probability"
          probability_threshold_attribute = 0.1
          s3_data_distribution_type = "ShardedByS3Key"
          s3_input_mode = "Pipe"
          start_time_offset = "-PT7H"
        },
        {
          data_captured_destination_s3_uri = "s3://captured-data-bucket"
          dataset_format = {
            json = {
              line = true
            }
          }
          local_path = "/mnt/endpoint"
          end_time_offset = "-PT7H"
          features_attribute = "feature"
          inference_attribute = "inference"
          probability_attribute = "probability"
          probability_threshold_attribute = 0.1
          s3_data_distribution_type = "ShardedByS3Key"
          s3_input_mode = "Pipe"
          start_time_offset = "-PT7H"
        }
      ]
      monitoring_output_config = {
        monitoring_outputs = [
          {
            s3_output = {
              local_path = "/monitor/output"
              s3_uri = "s3://monitor-output-bucket"
              s3_upload_mode = "Continuous"
            }
          }
        ]
        kms_key_id = "alias/key_id"
      }
      monitoring_resources = {
        cluster_config = {
          instance_count = 2
          instance_type = "ml.t3.large"
          volume_size_in_gb = 100
          volume_kms_key_id = "alias/my_key"
        }
      }
      role_arn = "arn::1234567:iam:my_role"
      baseline_config = {
        baselining_job_name = "baselining"
        constraints_resource = {
          s3_uri = "s3://my-constraint/constraint.csv"
        }
        statistics_resource = {
          s3_uri = "s3://my-statistics/statistics.csv"
        }
      }
      environment = {
        foo = "bar"
      }
      network_config = {
        enable_inter_container_traffic_encryption = true
        enable_network_isolation = true
        vpc_config = {
          SecurityGroupIds = ["sg-123456789"]
          Subnets = ["subnet-123456789"]
        }
      }
      monitoring_stopping_condition = {
        max_runtime_in_seconds = 3600
      }
    }
    monitoring_job_name = "my_job"
    type = "DataQuality"
    schedule_config = {
      schedule_expression = "cron(0 * ? * * *)"
    }   
  }
  name = "my_schedule"
  tags = ["tag1", "tag2"]
}

References

Sagemaker CreateMonitoringSchedule
Sagemaker DeleteMonitoringSchedule
Sagemaker DescribeMonitoringSchedule
Sagemaker ListMonitoringSchedules
Sagemaker StartMonitoringSchedule
Sagemaker StopMonitoringSchedule
Sagemaker UpdateMonitoringSchedule

Would you like to implement a fix?

Yes

[github-actions]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.
If you have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.