mlm_insights.builder.utils package¶
Submodules¶
mlm_insights.builder.utils.builder_utils module¶
- mlm_insights.builder.utils.builder_utils.get_feature_list(transformed_schema: Schema) List[str] ¶
- mlm_insights.builder.utils.builder_utils.get_transformed_input_schema(input_schema: Dict[str, FeatureType], transformers: List[Transformer]) Schema ¶
- mlm_insights.builder.utils.builder_utils.validate_feature_schema(transformed_schema: Schema) None ¶
This method validates the transformed schema to ensure the schema doesn’t have any discrepancies
- mlm_insights.builder.utils.builder_utils.validate_metrics(transformed_schema: Schema, metrics: MetricDetail) None ¶
- mlm_insights.builder.utils.builder_utils.validate_required_components(input_schema: Dict[str, FeatureType], reader: DataReader | None, data_frame: Any) None ¶
mlm_insights.builder.utils.generate_input_schema_utils module¶
- mlm_insights.builder.utils.generate_input_schema_utils.classify_variable_type(data: DataFrame, columns: List[str], dtype: str) Dict[str, Any] ¶
Classify the variable_type of feature based on dtype
Parameters¶
data: DataFrame columns: List[str]
List of feature name
- dtype: List[str]
List of dtype_types for each feature
Returns¶
Dict[str, Any]
- mlm_insights.builder.utils.generate_input_schema_utils.convert_column_type(column: str, target_features: List[str], prediction_features: List[str], prediction_score_features: List[str]) ColumnType ¶
Convert column_type of feature
Parameters¶
- column: str
Name of feature
- target_features: List[str]
List of target_features name
- prediction_features: List[str]
List of prediction_features name
- prediction_score_features: List[str]
List of prediction_score_features name
Returns¶
ColumnType
- mlm_insights.builder.utils.generate_input_schema_utils.convert_data_type(dtype: str) DataType ¶
Convert dtype of feature
Parameters¶
dtype: str data type of feature
Returns¶
DataType
- mlm_insights.builder.utils.generate_input_schema_utils.convert_variable_type(variable_type: str) VariableType ¶
Convert variable_type of feature
Parameters¶
variable_type: str variable_type of feature
Returns¶
VariableType
- mlm_insights.builder.utils.generate_input_schema_utils.create_output_schema(column_names: List[str], inferred_types: List[str], variable_type: Dict[str, Any], target_features: List[str], prediction_features: List[str], prediction_score_features: List[str]) Dict[str, Any] ¶
Create input_schema
Parameters¶
- column_names: List[str]
List of column_names name
- inferred_types: List[str]
List of inferred_types for each feature
- variable_type: List[str]
List of variable_type for each feature
- target_features: List[str]
List of target_features name
- prediction_features: List[str]
List of prediction_features name
- prediction_score_features: List[str]
List of prediction_score_features name
Returns¶
input_schema: Dict[str, Any]
- mlm_insights.builder.utils.generate_input_schema_utils.generate_input_schema_using_dataset(file_location: str, target_features: List[str], prediction_features: List[str], prediction_score_features: List[str]) Dict[str, Any] ¶
Method to generate the approximated input_schema based on the dataset .
Parameters¶
- file_locationstr
Sample dataset location
- target_featuresList[str]
List of target_features names
- prediction_featuresList[str]
List of target_features names
- prediction_score_featuresList[str]
List of target_features names
Returns¶
input_schema : Dict[str, Any]