You are viewing documentation for version 2 of the AWS SDK for Ruby. Version 3 documentation can be found here.
Class: Aws::SageMaker::Types::CreateAutoMLJobRequest
- Inherits:
-
Struct
- Object
- Struct
- Aws::SageMaker::Types::CreateAutoMLJobRequest
- Defined in:
- (unknown)
Overview
When passing CreateAutoMLJobRequest as input to an Aws::Client method, you can use a vanilla Hash:
{
auto_ml_job_name: "AutoMLJobName", # required
input_data_config: [ # required
{
data_source: { # required
s3_data_source: { # required
s3_data_type: "ManifestFile", # required, accepts ManifestFile, S3Prefix
s3_uri: "S3Uri", # required
},
},
compression_type: "None", # accepts None, Gzip
target_attribute_name: "TargetAttributeName", # required
},
],
output_data_config: { # required
kms_key_id: "KmsKeyId",
s3_output_path: "S3Uri", # required
},
problem_type: "BinaryClassification", # accepts BinaryClassification, MulticlassClassification, Regression
auto_ml_job_objective: {
metric_name: "Accuracy", # required, accepts Accuracy, MSE, F1, F1macro, AUC
},
auto_ml_job_config: {
completion_criteria: {
max_candidates: 1,
max_runtime_per_training_job_in_seconds: 1,
max_auto_ml_job_runtime_in_seconds: 1,
},
security_config: {
volume_kms_key_id: "KmsKeyId",
enable_inter_container_traffic_encryption: false,
vpc_config: {
security_group_ids: ["SecurityGroupId"], # required
subnets: ["SubnetId"], # required
},
},
},
role_arn: "RoleArn", # required
generate_candidate_definitions_only: false,
tags: [
{
key: "TagKey", # required
value: "TagValue", # required
},
],
}
Instance Attribute Summary collapse
-
#auto_ml_job_config ⇒ Types::AutoMLJobConfig
Contains CompletionCriteria and SecurityConfig.
-
#auto_ml_job_name ⇒ String
Identifies an Autopilot job.
-
#auto_ml_job_objective ⇒ Types::AutoMLJobObjective
Defines the objective of a an AutoML job.
-
#generate_candidate_definitions_only ⇒ Boolean
Generates possible candidates without training a model.
-
#input_data_config ⇒ Array<Types::AutoMLChannel>
Similar to InputDataConfig supported by Tuning.
-
#output_data_config ⇒ Types::AutoMLOutputDataConfig
Similar to OutputDataConfig supported by Tuning.
-
#problem_type ⇒ String
Defines the kind of preprocessing and algorithms intended for the candidates.
-
#role_arn ⇒ String
The ARN of the role that is used to access the data.
-
#tags ⇒ Array<Types::Tag>
Each tag consists of a key and an optional value.
Instance Attribute Details
#auto_ml_job_config ⇒ Types::AutoMLJobConfig
Contains CompletionCriteria and SecurityConfig.
#auto_ml_job_name ⇒ String
Identifies an Autopilot job. Must be unique to your account and is case-insensitive.
#auto_ml_job_objective ⇒ Types::AutoMLJobObjective
Defines the objective of a an AutoML job. You provide a AutoMLJobObjective$MetricName and Autopilot infers whether to minimize or maximize it. If a metric is not specified, the most commonly used ObjectiveMetric for problem type is automaically selected.
#generate_candidate_definitions_only ⇒ Boolean
Generates possible candidates without training a model. A candidate is a combination of data preprocessors, algorithms, and algorithm parameter settings.
#input_data_config ⇒ Array<Types::AutoMLChannel>
Similar to InputDataConfig supported by Tuning. Format(s) supported: CSV. Minimum of 500 rows.
#output_data_config ⇒ Types::AutoMLOutputDataConfig
Similar to OutputDataConfig supported by Tuning. Format(s) supported: CSV.
#problem_type ⇒ String
Defines the kind of preprocessing and algorithms intended for the candidates. Options include: BinaryClassification, MulticlassClassification, and Regression.
Possible values:
- BinaryClassification
- MulticlassClassification
- Regression
#role_arn ⇒ String
The ARN of the role that is used to access the data.
#tags ⇒ Array<Types::Tag>
Each tag consists of a key and an optional value. Tag keys must be unique per resource.