mlm_insights.builder package¶
Subpackages¶
- mlm_insights.builder.constants package
- mlm_insights.builder.exceptions package
- mlm_insights.builder.runner package
- mlm_insights.builder.usage package
- mlm_insights.builder.utils package
Submodules¶
mlm_insights.builder.builder_component module¶
- class mlm_insights.builder.builder_component.EngineDetail(engine_name: str | mlm_insights.constants.types.ExecutionEngineType, engine_client: Any = None)¶
Bases:
object
- engine_client: Any = None¶
- engine_name: str | ExecutionEngineType¶
Execution engine client created and passed by application. Insights uses this client, if passed.
- class mlm_insights.builder.builder_component.MetricDetail(univariate_metric: Dict[str, List[mlm_insights.core.metrics.metric_metadata.MetricMetadata]], dataset_metrics: List[mlm_insights.core.metrics.metric_metadata.MetricMetadata] | NoneType)¶
Bases:
object
- dataset_metrics: List[MetricMetadata] | None¶
- univariate_metric: Dict[str, List[MetricMetadata]]¶
- class mlm_insights.builder.builder_component.RunResult(profile: mlm_insights.core.profiles.profile.Profile, test_results: Union[mlm_insights.tests.test_results.TestResults, NoneType], post_processor_run_result: Dict[str, mlm_insights.core.post_processors.post_processor_result.PostProcessorResult] = <factory>)¶
Bases:
object
- post_processor_run_result: Dict[str, PostProcessorResult]¶
- test_results: TestResults | None¶
mlm_insights.builder.insights_builder module¶
- class mlm_insights.builder.insights_builder.InsightsBuilder¶
Bases:
object
Builder class for MLInsights Library. This will be used to build a Runner class instance using the builder components.
- build() Runner ¶
Method to build and return the Runner object and validate if required parameters are provided to create a Runner object.
Returns¶
- Runner
An Instance of Runner.
- with_data_frame(data_frame: Any) InsightsBuilder ¶
Method to set the data frame in the builder object.
Parameters¶
- data_frameAny
Data Frame object.
Returns¶
- InsightsBuilder
An Instance of InsightsBuilder.
- with_engine(engine: EngineDetail) InsightsBuilder ¶
Method to set the engine detail in the builder object.
Parameters¶
- engineEngineDetail
EngineDetail object type.
Returns¶
- InsightsBuilder
An Instance of InsightsBuilder.
- with_input_schema(input_schema: Dict[str, FeatureType]) InsightsBuilder ¶
Method to set the schema of the input data in the builder object.
Parameters¶
- input_schemaDict[str, FeatureType]
Dictionary of feature name string as key and feature type object as value.
Returns¶
- InsightsBuilder
An Instance of InsightsBuilder.
- with_input_schema_using_dataset(dataset_location: str, target_features: List[str] = [], prediction_features: List[str] = [], prediction_score_features: List[str] = []) InsightsBuilder ¶
Method to set the approximated input_schema based on the dataset in the builder object.
Parameters¶
- dataset_locationstr
The location of sample dataset. It supports local file storage url only.
- target_featuresList[str]
List of target_features names
- prediction_featuresList[str]
List of prediction_features names
- prediction_score_featuresList[str]
List of prediction_score_features names
Returns¶
- InsightsBuilder
An Instance of InsightsBuilder.
- with_metrics(metrics: MetricDetail) InsightsBuilder ¶
Method to set the MetricDetail for the feature in the builder object.
Parameters¶
- metricsMetricDetail
MetricDetail object type.
Returns¶
- InsightsBuilder
An Instance of InsightsBuilder.
- with_post_processors(post_processors: List[PostProcessor]) InsightsBuilder ¶
Method to set the list of post processors in the builder object.
Parameters¶
- post_processorsList[PostProcessor]
List of post processors.
Returns¶
- InsightsBuilder
An Instance of InsightsBuilder.
- with_reader(reader: DataReader) InsightsBuilder ¶
Method to set the reader in the builder object.
Parameters¶
- readerDataReader
Reader object instance.
Returns¶
- InsightsBuilder
An Instance of InsightsBuilder.
- with_reference_profile(profile: Profile | ProfileReader | None) InsightsBuilder ¶
Method to set the reference profile in the builder object, either by passing the profile directly or by specifying a Profile reader. Profile reader is used to read the reference profile for the run.
Parameters¶
profile : Profile or ProfileReader
Returns¶
- InsightsBuilder
An Instance of InsightsBuilder.
- with_tags(tags: Tags) InsightsBuilder ¶
Method to set the tags of a Profile.
Parameters¶
tags : Tags
Returns¶
- InsightsBuilder
An Instance of InsightsBuilder.
- with_test_config(test_config: TestConfig) InsightsBuilder ¶
Method to set the tests config in the builder object. Test config contains a list of test and user defined tags.
Parameters¶
test_config : TestConfig
Returns¶
- InsightsBuilder
An Instance of InsightsBuilder.
- with_transformers(transformers: List[Transformer]) InsightsBuilder ¶
Method to set the list of transformers in the builder object.
Parameters¶
- transformersList[Transformer]
List of transformers.
Returns¶
- InsightsBuilder
An Instance of InsightsBuilder.
mlm_insights.builder.schema_helper module¶
- class mlm_insights.builder.schema_helper.SchemaHelper(input_schema: Schema, quantitative_features: List[Field] | None, categorical_features: List[Field] | None)¶
Bases:
object
Class to provide Insights schema related APIs
- are_prediction_target_types_same() bool ¶
- get_categorical_features(max_features_count: int) List[str] ¶
- get_prediction_features() List[Field] ¶
- get_prediction_score_feature() List[Field] ¶
- get_prediction_types() Dict[Any, Any] | None ¶
- get_quantitative_features(max_features_count: int) List[str] ¶
- get_target_features() List[Field] ¶
- get_target_types() Dict[Any, Any] | None ¶
- has_prediction_score_column() bool ¶
- is_not_binary_or_multiclass() bool ¶
- is_valid_classification_types() bool ¶
- is_valid_regression_types() bool ¶
- static to_insights_schema(pa_schema: Schema) Dict[str, FeatureType] ¶
- mlm_insights.builder.schema_helper.map_to_name(field: Field) str ¶