Skip to content

ndarray Module

This section documents the NumPy ndarray utilities of the nextmv-scikit-learn package.

ndarray

NumPy ndarray wrapper for Pydantic model integration.

This module provides utilities to handle NumPy arrays in Pydantic models.

Variables

ndarray : Annotated A type annotation for NumPy arrays in Pydantic models.

ndarray module-attribute 

ndarray = Annotated[
    ndarray,
    BeforeValidator(lambda x: x),
    PlainSerializer(lambda x: tolist()),
]

ndarray: An annotated type that represents a NumPy array.

You can import the ndarray type directly from nextmv_sklearn:

from nextmv_sklearn import ndarray

This type is designed for use with Pydantic models to properly handle NumPy arrays for validation and serialization.

The BeforeValidator ensures the input is preserved as is, while PlainSerializer converts the ndarray to a Python list when serializing.

Examples:

>>> from pydantic import BaseModel
>>> from nextmv_sklearn import ndarray
>>> import numpy as np
>>>
>>> class MyModel(BaseModel):
...     data: ndarray
...
>>> model = MyModel(data=np.array([1, 2, 3]))
>>> model.model_dump_json()
'{"data":[1,2,3]}'