> ## Documentation Index
> Fetch the complete documentation index at: https://aidflow.aigensys.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Predictor.woe_stabilize

**def Predictor.woe\_stabilize**(self, data=None, encode=False) [\[source\]](https://github.com/aigensys-labs/aidflow/blob/main/AID/\TRUENAS\aig_modeling_Products\AID\1_0_Lib\AID\predictor\predictor.py#L458)

Apply trained WOE stability filtering.

Removes WOE features that were flagged as unstable
(distribution shift) during training.

#### Parameters

**`data`** : *`pDF | None`*
WOE-encoded data. If None, returns the list of\
stable WOE features.

**`encode`** : *`bool`*
If True, applies `woe_encode` first (expects\
binned data). If False, expects WOE-encoded input.

#### Returns

*`WoeStabilizerResult | list`*
• data provided: data subset to stable WOE features only\
• data is None: list of stable WOE feature names


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.