• Written in C and takes advantage of AVX2 for speed
• The fast implementation is tailored for contiguous 1-dimensional float32 arrays

$ pip install numpy-rms

import numpy_rms
import numpy as np

arr = np.arange(40, dtype=np.float32)
rms_series = numpy_rms.rms(arr, window_size=10)
print(rms_series.shape)  # (4,)

See CHANGELOG.md

• Install dev/build/test dependencies as denoted in pyproject.toml
• CC=clang pip install -e .
• pytest

This library is maintained/backed by Nomono, a Norwegian audio AI startup.