_wavelets#

cultivars.engine._core._wavelets = <module 'cultivars.engine._core._wavelets' from '/home/runner/work/cultivars/cultivars/src/cultivars/engine/_core/_wavelets.py'>[source]#

MODWT pyramid, multiresolution synthesis, Morlet transform, smoothing.

The maximal-overlap discrete wavelet transform keeps T coefficients at every level (no decimation), which makes it shift-invariant and lets the details of a multiresolution analysis line up with the data date by date; its price is redundancy. The continuous Morlet transform is what cross-wavelet coherence needs: complex, so phase is available, and smooth in scale. Filters are Percival and Walden’s tables.

Type:

Wavelet primitives