MLMachine

class zvt.ml.ml.MLMachine(entity_ids: List[str] = None, start_timestamp: str | Timestamp = '2015-01-01', end_timestamp: str | Timestamp = '2021-12-01', predict_start_timestamp: str | Timestamp = '2021-06-01', predict_steps: int = 20, level: IntervalLevel | str = IntervalLevel.LEVEL_1DAY, adjust_type: AdjustType | str = None, data_provider: str = None, label_method: str = 'raw')

Bases: object

__init__(entity_ids: List[str] = None, start_timestamp: str | Timestamp = '2015-01-01', end_timestamp: str | Timestamp = '2021-12-01', predict_start_timestamp: str | Timestamp = '2021-06-01', predict_steps: int = 20, level: IntervalLevel | str = IntervalLevel.LEVEL_1DAY, adjust_type: AdjustType | str = None, data_provider: str = None, label_method: str = 'raw') → None

Parameters:

• entity_ids

• start_timestamp

• end_timestamp

• predict_start_timestamp

• predict_steps

• level

• adjust_type

• data_provider

• label_method – raw, change, or behavior_cls

build_feature(entity_ids: List[str], start_timestamp: Timestamp, end_timestamp: Timestamp) → DataFrame

result df format

col1 col2 col3 …

entity_id timestamp

1.2 0.5 0.3 … 1.0 0.7 0.2 …

Parameters:

• entity_ids – entity id list

• start_timestamp

• end_timestamp

Return type:

pd.DataFrame