"""
This module provides a K-Nearest Neighbors (KNN) model wrapper, inheriting
from `aiqclib.common.base.scikit_learn_model_base.SklearnModelBase`.
It facilitates training, prediction, and evaluation of a KNN classifier using
Polars DataFrames, converting them to Pandas for compatibility with the
`sklearn` library.
"""
from typing import Dict, Any
from sklearn.neighbors import KNeighborsClassifier as SklearnKNN
from aiqclib.common.base.config_base import ConfigBase
from aiqclib.common.base.scikit_learn_model_base import SklearnModelBase
[docs]
class KNearestNeighbors(SklearnModelBase):
"""
A K-Nearest Neighbors (KNN) model wrapper class for training and testing.
Inherits from :class:`SklearnModelBase` to reuse common Scikit-Learn API logic.
Features include:
- Automatic application of ``model_params`` from the YAML config, if defined;
otherwise, uses default hyperparameters.
- Uses ``sklearn.neighbors.KNeighborsClassifier``.
.. note::
This class sets :attr:`expected_class_name` to ``"KNearestNeighbors"``.
Note that KNN can be computationally expensive during prediction for large datasets.
"""
expected_class_name: str = "KNearestNeighbors"
short_name: str = "KNN"
def __init__(self, config: ConfigBase) -> None:
"""
Initialize the KNN model with default or user-specified parameters.
:param config: A configuration object providing model parameters.
:type config: ConfigBase
"""
super().__init__(config=config)
self.model_params: Dict[str, Any] = {
"n_neighbors": 5,
"weights": "uniform",
"algorithm": "auto",
"leaf_size": 30,
"p": 2, # Euclidean distance
"metric": "minkowski",
"n_jobs": -1,
}
# Update model parameters with config step parameters
model_params = self.config.get_model_params(
self.expected_class_name, self.short_name
)
self.model_params.update(model_params)
self.allow_na = False
def _get_model_class(self) -> Any:
"""
Return the Scikit-Learn KNeighborsClassifier class.
:returns: The KNeighborsClassifier class.
:rtype: Any
"""
return SklearnKNN