Francesco Andreuzzi
Francesco Andreuzzi
We have a tutorial for `HODMD`, but not for `HankelDMD` (which is a simpler variant of DMD wrt `HODMD`).
It's not mandatory to have the same number of time snapshots in the dataset used for training the DMDs in the the partitioned version of `ParametricDMD`. We should state this...
At the moment applying methods in `dmd_modes_tuner` to `ParametricDMD` is a little bit cumbersome. The improved code should apply automatically the same transformation to all the DMDs inside `ParametricDMD`.
After merging #241 there are still some DMD variants that lack the parameter to control Tikhonov regularization. For instance: + `hankeldmd.py` + `hodmd.py` + `spdmd.py`
Every time a new subclass of `DMDBase` is created there is a lot of recurrent documentation to be copied from another instance to the new one, especially that for `__init__()`....
3193929266658b5f01f5650e69e225c02433f66d, 480bcb451b26d11579c904b1df3b5138248d3501, 2f757b3c349235a46206359e8cdb50ca5a8290d1 introduced several tools for the automatic selection of DMD modes. However the way those tools are intended to be used is not explained in a tutorial at...
The link https://networkx.org/grave/stable/ leads to a 404 error.
GLAPPO
This PR showcases a proof of concept for generic linear algebra in PyDMD (`cdmd.py` has been adapted to the new framework). At the moment the supported backends are PyTorch and...
Very low priority, but we should replace with enums all the existing parameter which use many constant strings to differentiate among available options. This would improve user experience since they...
https://peps.python.org/pep-0698/ In order to support old versions we could use the one provided by `typing_extensions`.