FatResConBUT/TT-DMD

TT-DMD Online Solver

Tensor-Train Dynamic Mode Decomposition — interactive solver for spatiotemporal prediction.

Developed at the Institute of Structural Mechanics, Brno University of Technology (VUT) as part of the ERC Synergy Grant project FatResCon (Grant ID: 101167045).

Features

• —TT-DMD — Tensor-Train DMD for high-dimensional spatiotemporal tensors
• —Classic DMD — Standard SVD-based DMD for 2D data
• —Interactive Eigenvalue Spectrum (complex plane, unit circle)
• —Growth / Frequency Map of DMD modes
• —Relative L1 Error over time (training vs prediction split)
• —Supports .npy, .mat, .csv input formats
• —Extended prediction & export

Demo Data

📁 Download demo test data (Google Drive)

Citation

If you use this solver in research, please cite:

"Exponential Data Compression via Tensor-Train Dynamic Mode Decomposition for Predicting Concrete Deterioration"

Contact

• —Yue Li — yue.li@vut.cz
• —Miroslav Vořechovský — miroslav.vorechovsky@vut.cz

License

MIT © 2026 Yue Li, Miroslav Vořechovský