MODULE REFERENCE

dataframe

anomx.data.processing.dataframe

Composition of fitted preprocessing steps.

ProcessingPipeline

ProcessingPipeline(steps: Sequence[Transform])

Bases: Transform

Fit transforms sequentially on training data and reuse them for inference.

Steps are owned by the caller and retain their fitted state. Frames are supplied explicitly, so applying a pipeline never triggers remote dataset loading. An empty pipeline is an identity transform that returns a copy.

Signature parameters
Parameter	Type	Default
`steps`	`Sequence[Transform]`	`required`

Members

fitmethod

# ProcessingPipeline.fit Source ↗

fit(frame: pd.DataFrame) -> Self

Signature parameters
Parameter	Type	Default
`frame`	`pd.DataFrame`	`required`

Returns Self

fit_transformmethod

# ProcessingPipeline.fit_transform Source ↗

fit_transform(frame: pd.DataFrame) -> pd.DataFrame

Signature parameters
Parameter	Type	Default
`frame`	`pd.DataFrame`	`required`

Returns pd.DataFrame

transformmethod

# ProcessingPipeline.transform Source ↗

transform(frame: pd.DataFrame) -> pd.DataFrame

Signature parameters
Parameter	Type	Default
`frame`	`pd.DataFrame`	`required`

Returns pd.DataFrame

transform_batchesmethod

# ProcessingPipeline.transform_batches Source ↗

transform_batches(batches: Iterator[pd.DataFrame]) -> Iterator[pd.DataFrame]

Signature parameters
Parameter	Type	Default
`batches`	`Iterator[pd.DataFrame]`	`required`

Returns Iterator[pd.DataFrame]