> ## 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.

# ObservationStabilizerWorker.transform

**def ObservationStabilizerWorker.transform**(self, data) [\[source\]](https://github.com/aigensys-labs/aidflow/blob/main/AID/\TRUENAS\aig_modeling_Products\AID\1_0_Lib\AID\data\screen\obs_stabilizer.py#L553)

Apply the trained outlier filter to new data.

#### Parameters

**`data`** : *`pDF`*
Input data to filter.

#### Returns

*`ObsStabilizerResult`*
Cleaned data (outliers removed) + outlier flags.\
As the final step of the Screening phase, returned `data` is converted to native\
Pandas/NumPy dtypes. This single-conversion boundary minimizes peak pipeline\
memory footprint by preventing redundant downstream allocations and ensuring\
full compatibility with numpy, LightGBM, XGBoost, and Scikit-Learn.


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