yeom

yeom is a collection of simple utilities for statistics, machine learning, design of experiments, and image processing.

Installation

Install the package from the repository root:

python -m pip install .

To install the package in editable mode with its test dependencies:

python -m pip install -e ".[test]"

Python 3.9 or later is required.

Converting an OpenCV image to Pillow

to_pil converts an OpenCV image array to a PIL.Image.Image object. OpenCV is not installed as a dependency of yeom; install exactly one OpenCV package appropriate for your environment if you need it to load images.

import cv2 as cv

from yeom import to_pil

cv_image = cv.imread("sample.png", cv.IMREAD_UNCHANGED)
if cv_image is None:
    raise FileNotFoundError("sample.png")

pil_image = to_pil(cv_image)
pil_image.show()

The conversion depends on the shape of the input array:

For arrays whose dtype is not uint8, values are clipped to the range 0–255 and then converted to uint8. Arrays with unsupported dimensions or channel counts raise ValueError.

Design-of-experiments utilities

yeom.doe includes utilities for calculating variance inflation factors, estimating regression-coefficient power and required sample sizes, and generating D-optimal designs.

from yeom import doe

design, determinant = doe.coordinate_exchange(4, "A + B")
print(design)
print(determinant)

Testing

pytest

License

MIT License