openfluids is a home for small, focused, open packages built around dynamical
systems, chaos, and the spectral analysis of spatiotemporal data. Each tool does
one thing, and is meant to be read, reused, and combined.
fftkit — spectra of physical
signals, and one FFT API over eight backends (SciPy, NumPy, Intel MKL, CuPy,
PyTorch, TensorFlow, pyFFTW, Accelerate). spectrum() handles the resampling,
detrending and windowing a non-uniformly sampled record needs, and reports what
it did; below that, swapping backend= leaves your call sites alone.
openmodalpy — nine modal
decomposition methods for spatiotemporal data behind one API: POD, mPOD,
PSD-POD, SPOD, ST-POD, DMD/HODMD and BSMD, driven by a config file or a CLI.
Pure NumPy/SciPy, no external solvers.
dynachaos — dynamical-systems
analysis of measured or simulated time signals: maps and coupled-map lattices,
Lyapunov exponents, recurrence quantification, entropy diagnostics, correlation
dimension and multifractal spectra, with Rust-accelerated kernels where those
have parity tests.
dsgbr — spectral peak detection for
noisy power spectra that slope over several decades, using a Savitzky–Golay
search scored against a rolling-median baseline rather than a fixed threshold.
chaos-atlas — an interactive
explorer for chaotic dynamical systems: strange attractors, bifurcations,
Lyapunov exponents, fractals and coupled-map-lattice patterns. Runs in the
browser; the map kernels also ship as a Python package.
They are built to compose: openmodalpy runs its transforms through fftkit, so
a backend installed once is available to everything above it.
More tools will appear here as they are released.