sagemaker.mlops.workflow.quality_check_step

The step definitions for workflow.

Classes

DataQualityCheckConfig(baseline_dataset, ...)	Data Quality Check Config.
ModelQualityCheckConfig(baseline_dataset, ...)	Model Quality Check Config.
QualityCheckConfig(baseline_dataset, ...[, ...])	Quality Check Config.
QualityCheckStep(name, quality_check_config, ...)	QualityCheck step for workflow.

class sagemaker.mlops.workflow.quality_check_step.DataQualityCheckConfig(baseline_dataset: str | PipelineVariable, dataset_format: dict, record_preprocessor_script: str | None = None, *, output_s3_uri: str | PipelineVariable | None = None, post_analytics_processor_script: str | None = None)

Bases: QualityCheckConfig

Data Quality Check Config.

record_preprocessor_script

The path to the record preprocessor script (default: None). This can be a local path or an S3 uri string but CANNOT be any type of the PipelineVariable.

Type:

str

record_preprocessor_script: str

class sagemaker.mlops.workflow.quality_check_step.ModelQualityCheckConfig(baseline_dataset: str | PipelineVariable, dataset_format: dict, problem_type: str | PipelineVariable, inference_attribute: str | PipelineVariable | None = None, probability_attribute: str | PipelineVariable | None = None, ground_truth_attribute: str | PipelineVariable | None = None, probability_threshold_attribute: str | PipelineVariable | None = None, *, output_s3_uri: str | PipelineVariable | None = None, post_analytics_processor_script: str | None = None)

Bases: QualityCheckConfig

Model Quality Check Config.

problem_type

The type of problem of this model quality monitoring. Valid values are “Regression”, “BinaryClassification”, “MulticlassClassification”.

Type:

str or PipelineVariable

inference_attribute

Index or JSONpath to locate predicted label(s) (default: None).

Type:

str or PipelineVariable

probability_attribute

Index or JSONpath to locate probabilities (default: None).

Type:

str or PipelineVariable

ground_truth_attribute (str or PipelineVariable

Index or JSONpath to locate actual label(s) (default: None).

probability_threshold_attribute

Threshold to convert probabilities to binaries (default: None).

Type:

str or PipelineVariable

ground_truth_attribute: str | PipelineVariable

inference_attribute: str | PipelineVariable

probability_attribute: str | PipelineVariable

probability_threshold_attribute: str | PipelineVariable

problem_type: str | PipelineVariable

class sagemaker.mlops.workflow.quality_check_step.QualityCheckConfig(baseline_dataset: str | PipelineVariable, dataset_format: dict, *, output_s3_uri: str | PipelineVariable | None = None, post_analytics_processor_script: str | None = None)

Bases: ABC

Quality Check Config.

baseline_dataset

The path to the baseline_dataset file. This can be a local path or an S3 uri string

Type:

str or PipelineVariable

dataset_format

The format of the baseline_dataset.

Type:

dict

output_s3_uri

Desired S3 destination of the constraint_violations and statistics json files (default: None). If not specified an auto generated path will be used: “s3://<default_session_bucket>/model-monitor/baselining/<job_name>/results”

Type:

str or PipelineVariable

post_analytics_processor_script

The path to the record post-analytics processor script (default: None). This can be a local path or an S3 uri string but CANNOT be any type of the PipelineVariable.

Type:

str

baseline_dataset: str | PipelineVariable

dataset_format: dict

output_s3_uri: str | PipelineVariable

post_analytics_processor_script: str

class sagemaker.mlops.workflow.quality_check_step.QualityCheckStep(name: str, quality_check_config: QualityCheckConfig, check_job_config: CheckJobConfig, skip_check: bool | PipelineVariable = False, fail_on_violation: bool | PipelineVariable = True, register_new_baseline: bool | PipelineVariable = False, model_package_group_name: str | PipelineVariable | None = None, supplied_baseline_statistics: str | PipelineVariable | None = None, supplied_baseline_constraints: str | PipelineVariable | None = None, display_name: str | None = None, description: str | None = None, cache_config: CacheConfig | None = None, depends_on: List[str | Step] | None = None)

Bases: Step

QualityCheck step for workflow.

property arguments: Dict[str, Any] | List[Dict[str, Any]]

The arguments dict that is used to define the QualityCheck step.

property properties

A Properties object representing the output parameters of the QualityCheck step.

to_request() → Dict[str, Any] | List[Dict[str, Any]]

Updates the dictionary with cache configuration etc.