logml.eda

Classes

EDAArtifactsOutputStructure(*args, **kwargs)

Defines outputs structure for EDA artifacts.

class logml.eda.EDAArtifactsOutputStructure(*args, **kwargs)

Bases: logml.common.OutputStructure

Defines outputs structure for EDA artifacts.

- strata_id
  |
  |- eda
     |-artifacts
       |-metadata.pickle
       |-correlation.pickle
       ...
     |- input
        |- data.csv
        |- dataset_metadata.yaml

EDA_ROOT: str = 'eda'

ARTIFACTS: str = 'artifacts'

DATASET: str = 'input'

DATASET_NAME: str = 'data.csv'

DATASET_METADATA_NAME: str = 'dataset_metadata.yaml'

CG_FILE_NAME: str = 'corr_groups.csv'

property cg_file_path: pathlib.Path

Path to Correlation groups CSV file.

artifacts_exist() → bool

Checks whether EDA artifacts exist.

Modules

logml.eda.artifacts
logml.eda.artifacts_producers
logml.eda.manager
logml.eda.runner