koopman-dmd

Dynamic Mode Decomposition (DMD) with Koopman operator theory extensions, with a Rust core.

DMD extracts spatiotemporal coherent structures from time-series data, giving a linear operator that approximates the dynamics of a possibly nonlinear system. This package wraps the koopman-dmd Rust crate via PyO3, so the numerics run at native speed with no BLAS/LAPACK installation required.

Installation

pip install koopman-dmd

The distribution is named koopman-dmd; the import name is koopman_dmd:

import koopman_dmd

Prebuilt wheels are published for Linux (x86_64, aarch64), macOS (x86_64, arm64), and Windows (x86_64) on Python 3.9+. Installing from source requires a Rust toolchain.

Quick start

Data is passed as a NumPy array with one row per variable and one column per time step.

import numpy as np
import koopman_dmd

# A 2-variable oscillating signal, shape (2, 100)
t = np.linspace(0, 10, 100)
x = np.vstack([np.sin(t), np.cos(t)])

d = koopman_dmd.DMD(x, rank=2, dt=t[1] - t[0])

print(d.eigenvalues)      # (r, 2) array of [real, imag]
print(d.modes)            # DMD modes
print(d.singular_values)

# Forecast 10 steps beyond the input
future = d.predict(10)

# Per-mode frequency, growth rate, amplitude, and stability
for mode in d.spectrum():
    print(mode)

Note that the data is passed to the constructor — there is no separate fit() step.

`DMD`

koopman_dmd.DMD(x, rank=None, center=False, dt=1.0, lifting=None, lifting_param=None)

rank=None selects a truncation rank automatically (99% of variance). lifting enables Extended DMD and accepts "polynomial", "polynomial_cross", "trigonometric", or "delay", with the degree / harmonic count / delay count given by lifting_param.

Properties: rank, data_dim, center, dt, eigenvalues, modes, amplitudes, singular_values

Methods:

Method	Returns
`predict(n_ahead, x0=None, method="modes")`	forecast array; `method` is `"modes"` or `"matrix"`
`reconstruct(n_steps, modes_subset=None)`	reconstruction from all or selected modes
`spectrum()`	per-mode frequency, growth rate, amplitude, stability — uses the `dt` given at construction
`stability()`	`(is_stable, is_unstable, is_marginal, spectral_radius)`
`error()`	`(rmse, mae, mape, rel_err)`
`dominant_modes(n, criterion="amplitude")`	indices of the `n` most significant modes
`residual()`	`(absolute, relative)`

Extended DMD with lifting

d = koopman_dmd.DMD(x, lifting="polynomial", lifting_param=2)

`HankelDMD`

Time-delay embedding, for scalar signals or systems with few measured variables.

koopman_dmd.HankelDMD(y, delays=None, rank=None, dt=1.0)

Properties: rank, delays, n_obs, residual, eigenvalues Methods: predict(n_ahead), reconstruct(n_steps)

y = np.sin(np.linspace(0, 4 * np.pi, 200)).reshape(1, -1)
h = koopman_dmd.HankelDMD(y, delays=20)
print(h.rank)          # 2 — chosen automatically
print(h.predict(10))

Leaving rank=None is usually right. Requesting a rank higher than the signal actually supports (a pure sinusoid is rank 2) leaves the reduced operator near-singular, and the eigendecomposition can fail with NoConvergence.

`GLA`

Generalized Laplace Analysis — computes Koopman eigenfunctions directly via weighted time averages.

koopman_dmd.GLA(y, eigenvalues=None, n_eigenvalues=5, tol=1e-6, max_iter=None)

Properties: n_obs, n_time, eigenvalues, convergence, residuals Methods: predict(n_ahead), reconstruct(modes_to_use=None)

Phase space analysis

For area-preserving and chaotic maps.

Maps and their params keys — params is a dict, and omitting it uses defaults:

`map_name`	State dim	Parameters
`"standard"` (Chirikov)	2	`epsilon`
`"froeschle"`	4	`epsilon`, `eta`
`"extended_standard"`	3	`epsilon`, `delta`
`"henon"`	2	`a`, `b`
`"logistic"`	1	`r`

Observables: "identity", "sin_pi", "cos_pi", "sin_pi_xy", "cos_pi_xy", "sin_2pi", "cos_2pi", "trig_product"

Initial conditions are NumPy arrays, not lists.

ic = np.array([0.5, 0.3])

# Iterate a map from an initial condition
traj = koopman_dmd.generate_trajectory("standard", ic, 1000, {"epsilon": 0.12})

# Harmonic time average at a single initial condition
mag, phase, hta_re, hta_im = koopman_dmd.harmonic_time_average(
    "standard", ic, "sin_pi", 0.5, 10000, {"epsilon": 0.12}
)

# Mesochronic plot over a grid (parallelized in Rust).
# Returns (hta_magnitude, phase, x_coords, y_coords).
hta, phase, x_coords, y_coords = koopman_dmd.mesochronic_compute(
    "standard", (0.0, 1.0), (0.0, 1.0), 100, "sin_pi", 0.5, 10000, {"epsilon": 0.12}
)

# Classify orbits as regular, resonating, or chaotic from HTA magnitudes
labels = koopman_dmd.classify_phase_space(hta.ravel())

# Convergence history of the time average
conv = koopman_dmd.hta_convergence("standard", ic, "sin_pi", 0.5, 10000, {"epsilon": 0.12})

Other languages

Rust — koopman-dmd on crates.io
R — koopman.dmd (extendr)

Development

pip install maturin pytest numpy
maturin develop --release
pytest tests/

References

Schmid, P.J. (2010). Dynamic mode decomposition of numerical and experimental data. Journal of Fluid Mechanics, 656, 5–28.
Kutz, J.N., Brunton, S.L., Brunton, B.W., & Proctor, J.L. (2016). Dynamic Mode Decomposition: Data-Driven Modeling of Complex Systems. SIAM.
Mezić, I. (2020). Spectrum of the Koopman operator, spectral expansions in functional spaces, and state-space geometry. arXiv:2009.05883
Levnajić, Z. & Mezić, I. (2014). Ergodic theory and visualization. arXiv:0808.2182v2

License

MIT — see LICENSE.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

koopman_dmd-0.1.0.tar.gz (70.5 kB view details)

Uploaded Aug 6, 2026 Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

koopman_dmd-0.1.0-cp39-abi3-win_amd64.whl (980.1 kB view details)

Uploaded Aug 6, 2026 CPython 3.9+Windows x86-64

koopman_dmd-0.1.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.3 MB view details)

Uploaded Aug 6, 2026 CPython 3.9+manylinux: glibc 2.17+ x86-64

koopman_dmd-0.1.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl (866.1 kB view details)

Uploaded Aug 6, 2026 CPython 3.9+manylinux: glibc 2.17+ ARM64

koopman_dmd-0.1.0-cp39-abi3-macosx_11_0_arm64.whl (741.2 kB view details)

Uploaded Aug 6, 2026 CPython 3.9+macOS 11.0+ ARM64

koopman_dmd-0.1.0-cp39-abi3-macosx_10_12_x86_64.whl (1.1 MB view details)

Uploaded Aug 6, 2026 CPython 3.9+macOS 10.12+ x86-64

File details

Details for the file koopman_dmd-0.1.0.tar.gz.

File metadata

Download URL: koopman_dmd-0.1.0.tar.gz
Upload date: Aug 6, 2026
Size: 70.5 kB
Tags: Source
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for koopman_dmd-0.1.0.tar.gz
Algorithm	Hash digest
SHA256	`106b2d3f83b0b11d76bf035db18a8bd1b1238de250e617053509bf69d8f967d1`
MD5	`3837a64279524852b1459602c9442b50`
BLAKE2b-256	`b332cb298101cd69e6edf18f55b665f197311800af2aec137339a45ca2ea9fe4`

See more details on using hashes here.

Provenance

The following attestation bundles were made for koopman_dmd-0.1.0.tar.gz:

Publisher: release-python.yml on jimeharrisjr/rust-dmd

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: koopman_dmd-0.1.0.tar.gz
- Subject digest: 106b2d3f83b0b11d76bf035db18a8bd1b1238de250e617053509bf69d8f967d1
- Sigstore transparency entry: 2352462886
- Sigstore integration time: Aug 6, 2026
Source repository:
- Permalink: jimeharrisjr/rust-dmd@6e0455f8412c0f85dd086dab2b5cf6252a08bab7
- Branch / Tag: refs/tags/py-v0.1.0
- Owner: https://github.com/jimeharrisjr
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: release-python.yml@6e0455f8412c0f85dd086dab2b5cf6252a08bab7
- Trigger Event: push

File details

Details for the file koopman_dmd-0.1.0-cp39-abi3-win_amd64.whl.

File metadata

Download URL: koopman_dmd-0.1.0-cp39-abi3-win_amd64.whl
Upload date: Aug 6, 2026
Size: 980.1 kB
Tags: CPython 3.9+, Windows x86-64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for koopman_dmd-0.1.0-cp39-abi3-win_amd64.whl
Algorithm	Hash digest
SHA256	`76671f76c098f207a8be1c0cace8fd86ad061a2c1f7f29f0946f55fef23a8354`
MD5	`cfec4fa47c69339f4d797d22240c99a5`
BLAKE2b-256	`ac46a45cfe2fb0825e0d0f9dcfdcb8e13e0fa7c64277f2a81f30cab3a9b9979f`

See more details on using hashes here.

Provenance

The following attestation bundles were made for koopman_dmd-0.1.0-cp39-abi3-win_amd64.whl:

Publisher: release-python.yml on jimeharrisjr/rust-dmd

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: koopman_dmd-0.1.0-cp39-abi3-win_amd64.whl
- Subject digest: 76671f76c098f207a8be1c0cace8fd86ad061a2c1f7f29f0946f55fef23a8354
- Sigstore transparency entry: 2352463008
- Sigstore integration time: Aug 6, 2026
Source repository:
- Permalink: jimeharrisjr/rust-dmd@6e0455f8412c0f85dd086dab2b5cf6252a08bab7
- Branch / Tag: refs/tags/py-v0.1.0
- Owner: https://github.com/jimeharrisjr
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: release-python.yml@6e0455f8412c0f85dd086dab2b5cf6252a08bab7
- Trigger Event: push

File details

Details for the file koopman_dmd-0.1.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.

File metadata

Download URL: koopman_dmd-0.1.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Upload date: Aug 6, 2026
Size: 1.3 MB
Tags: CPython 3.9+, manylinux: glibc 2.17+ x86-64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for koopman_dmd-0.1.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm	Hash digest
SHA256	`32727c1d8fc94b3025d89b496b1f73f19e1c1044590d857db909b57f6381f674`
MD5	`e39acb9bac989c8261fd63acc3406746`
BLAKE2b-256	`4605ad04aa6c5cef330cfbd5f58cf6ed47c7e6cf15dfdf3311cfb86677f0a16e`

See more details on using hashes here.

Provenance

The following attestation bundles were made for koopman_dmd-0.1.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl:

Publisher: release-python.yml on jimeharrisjr/rust-dmd

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: koopman_dmd-0.1.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
- Subject digest: 32727c1d8fc94b3025d89b496b1f73f19e1c1044590d857db909b57f6381f674
- Sigstore transparency entry: 2352463105
- Sigstore integration time: Aug 6, 2026
Source repository:
- Permalink: jimeharrisjr/rust-dmd@6e0455f8412c0f85dd086dab2b5cf6252a08bab7
- Branch / Tag: refs/tags/py-v0.1.0
- Owner: https://github.com/jimeharrisjr
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: release-python.yml@6e0455f8412c0f85dd086dab2b5cf6252a08bab7
- Trigger Event: push

File details

Details for the file koopman_dmd-0.1.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl.

File metadata

Download URL: koopman_dmd-0.1.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
Upload date: Aug 6, 2026
Size: 866.1 kB
Tags: CPython 3.9+, manylinux: glibc 2.17+ ARM64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for koopman_dmd-0.1.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
Algorithm	Hash digest
SHA256	`f1b30cdcb93ad0e2c0e8d7d459465868b90fa763dafc4d2698f47a71e2f28a3e`
MD5	`e5ebbadcf6ff51e7a9be1964ce689b0d`
BLAKE2b-256	`c4c209c07bb8fc8c4c1330328d00821896fa168a1dbed7356d295d9b25321136`

See more details on using hashes here.

Provenance

The following attestation bundles were made for koopman_dmd-0.1.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl:

Publisher: release-python.yml on jimeharrisjr/rust-dmd

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: koopman_dmd-0.1.0-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
- Subject digest: f1b30cdcb93ad0e2c0e8d7d459465868b90fa763dafc4d2698f47a71e2f28a3e
- Sigstore transparency entry: 2352463283
- Sigstore integration time: Aug 6, 2026
Source repository:
- Permalink: jimeharrisjr/rust-dmd@6e0455f8412c0f85dd086dab2b5cf6252a08bab7
- Branch / Tag: refs/tags/py-v0.1.0
- Owner: https://github.com/jimeharrisjr
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: release-python.yml@6e0455f8412c0f85dd086dab2b5cf6252a08bab7
- Trigger Event: push

File details

Details for the file koopman_dmd-0.1.0-cp39-abi3-macosx_11_0_arm64.whl.

File metadata

Download URL: koopman_dmd-0.1.0-cp39-abi3-macosx_11_0_arm64.whl
Upload date: Aug 6, 2026
Size: 741.2 kB
Tags: CPython 3.9+, macOS 11.0+ ARM64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for koopman_dmd-0.1.0-cp39-abi3-macosx_11_0_arm64.whl
Algorithm	Hash digest
SHA256	`e7924c4516a2a04c1a68580e3f0720dfc1dac2461a5f026633ba69d01e185f27`
MD5	`43bd6800d3bb1598ac4c3451b44b2761`
BLAKE2b-256	`fda669ba175df1a8acb946a48f4fb2c42db7b7d49cfcfd3ddf8c682758e77534`

See more details on using hashes here.

Provenance

The following attestation bundles were made for koopman_dmd-0.1.0-cp39-abi3-macosx_11_0_arm64.whl:

Publisher: release-python.yml on jimeharrisjr/rust-dmd

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: koopman_dmd-0.1.0-cp39-abi3-macosx_11_0_arm64.whl
- Subject digest: e7924c4516a2a04c1a68580e3f0720dfc1dac2461a5f026633ba69d01e185f27
- Sigstore transparency entry: 2352463391
- Sigstore integration time: Aug 6, 2026
Source repository:
- Permalink: jimeharrisjr/rust-dmd@6e0455f8412c0f85dd086dab2b5cf6252a08bab7
- Branch / Tag: refs/tags/py-v0.1.0
- Owner: https://github.com/jimeharrisjr
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: release-python.yml@6e0455f8412c0f85dd086dab2b5cf6252a08bab7
- Trigger Event: push

File details

Details for the file koopman_dmd-0.1.0-cp39-abi3-macosx_10_12_x86_64.whl.

File metadata

Download URL: koopman_dmd-0.1.0-cp39-abi3-macosx_10_12_x86_64.whl
Upload date: Aug 6, 2026
Size: 1.1 MB
Tags: CPython 3.9+, macOS 10.12+ x86-64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for koopman_dmd-0.1.0-cp39-abi3-macosx_10_12_x86_64.whl
Algorithm	Hash digest
SHA256	`d16efd12fa7ca005da43bf3255aa09c0844d73c94eacf04f56b8f11f6aaca416`
MD5	`d16a80b81e7f4cd66a0eac232f2ec299`
BLAKE2b-256	`f492b8a061c22f7ed7e137840ce72d307864030dee28307a3043afd48fdfb087`

See more details on using hashes here.

Provenance

The following attestation bundles were made for koopman_dmd-0.1.0-cp39-abi3-macosx_10_12_x86_64.whl:

Publisher: release-python.yml on jimeharrisjr/rust-dmd

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: koopman_dmd-0.1.0-cp39-abi3-macosx_10_12_x86_64.whl
- Subject digest: d16efd12fa7ca005da43bf3255aa09c0844d73c94eacf04f56b8f11f6aaca416
- Sigstore transparency entry: 2352463172
- Sigstore integration time: Aug 6, 2026
Source repository:
- Permalink: jimeharrisjr/rust-dmd@6e0455f8412c0f85dd086dab2b5cf6252a08bab7
- Branch / Tag: refs/tags/py-v0.1.0
- Owner: https://github.com/jimeharrisjr
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: release-python.yml@6e0455f8412c0f85dd086dab2b5cf6252a08bab7
- Trigger Event: push

Release history Release notifications | RSS feed

This release

0.1.0

Aug 6, 2026

koopman-dmd 0.1.0

koopman-dmd

Installation

Quick start

DMD

Extended DMD with lifting

HankelDMD

GLA

Phase space analysis

Other languages

Development

References

License

Project links

Key dates

1 maintainer

Credits

GitHub Statistics

License expression

Requires

Provides Extra

Tags

Classifiers

Download files

Source Distribution

Built Distributions

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

Release history Release notifications | RSS feed

`DMD`

`HankelDMD`

`GLA`