PyDMD: A Python Package for Robust Dynamic Mode Decomposition

Authors: Sara M. Ichinaga, Francesco Andreuzzi, Nicola Demo, Marco Tezzele, Karl Lapo, Gianluigi Rozza, Steven L. Brunton, J. Nathan Kutz

JMLR 2024 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this paper, we introduce the version 1.0 release of Py DMD, which includes new data preprocessors, plotting tools, and a number of cutting-edge DMD methods specifically designed to handle real-world data that may be noisy, multiscale, parameterized, prohibitively high-dimensional, and even strongly nonlinear. ... Our code is unit tested, regularly maintained, and completely open-source under the MIT license. ... If we let X denote our data matrix, we can preprocess our data, perform DMD, and visualize the resulting spatiotemporal modes with the following code. ... Figure 1: Sample plot summary function output using fluid flow past a cylinder data with Reynolds number Re = 100. Data is available at dmdbook.com/DATA.zip.
Researcher Affiliation Academia 1 Department of Applied Mathematics, University of Washington, Seattle, WA 98195, USA 2 Mathematics Area, math Lab, SISSA, via Bonomea 265, I-34136 Trieste, Italy 3 Department of Mathematics, Emory University, Atlanta, GA 30322, USA 4 CERN, Geneva, Switzerland 5 Department of Atmospheric and Cryospheric Sciences, University of Innsbruck, Innsbruck, Austria 6 Department of Mechanical Engineering, University of Washington, Seattle, WA 98195, USA
Pseudocode Yes If we let X denote our data matrix, we can preprocess our data, perform DMD, and visualize the resulting spatiotemporal modes with the following code. ... 1 from pydmd import DMD 2 from pydmd.preprocessing import zero_mean_preprocessing 3 from pydmd.plotter import plot_summary 5 dmd = DMD(svd_rank =12) # Build DMD model. 6 dmd = zero_mean_preprocessing (dmd) # Wrap with data preprocessor. 7 dmd.fit(X) # Fit DMD model to snapshot data. 8 plot_summary(dmd) # Plot key spatiotemporal modes.
Open Source Code Yes The entire codebase is released under the MIT license and is available at https://github.com/Py DMD/Py DMD.
Open Datasets Yes Figure 1: Sample plot summary function output using fluid flow past a cylinder data with Reynolds number Re = 100. Data is available at dmdbook.com/DATA.zip.
Dataset Splits No No specific dataset split information (percentages, sample counts, or methodology for dividing data into training, validation, or test sets) is provided in the paper. The paper focuses on presenting a software package and its capabilities, demonstrating it with a sample fluid flow dataset without detailing experimental reproduction requirements for splits.
Hardware Specification No The paper describes a Python package and its features but does not provide any specific details about the hardware (e.g., GPU/CPU models, memory) used for development, testing, or example execution.
Software Dependencies No The paper lists several standard Python libraries (Num Py, Sci Py, Matplotlib, Scikit-learn) but does not specify their version numbers, which are required for reproducible software dependencies.
Experiment Setup Yes 5 dmd = DMD(svd_rank =12) # Build DMD model.