Marco Tezzele

@IvanPMorenoMarcos just change tend by adding the number of timesteps you want to predict in the future. For example if the train has 100 time instants and the test 20,...

You need to increase it. That dictionary tells you the starting time, the time step and the final time. You need to change only the final time (+20 for example)...

@benoitpaillard the assumption we made is that for prediction purposes you need to provide new inputs up to the chosen `t_end`. Up to now you have to provide also the...

Hi! You are right, up to now only a single matrix is available. Feel free to implement the modification with some small tests, and then we can review the code...

@Kolaru sure, you can propose such a solution and then we will see together whether there is a better integration or not ;) Contributions are more than welcome!

Dear @rajarshiroychowdhury, regularization is for sure a possible way to handle such problem. It is possible we will add this feature in the future. In the near term we are...

#108 addresses point 2. For what concerns point 1 I would add something in the `_col_major_2darray(X)` method. Are you aware of some literature doing regularization before applying DMD? In case...

Thanks. Can you tell me the shape of `mesh_points` after the following line: `mesh_points = handler.parse('../tests/test_datasets/test_pipe.k')` Moreover do you have the same problem using the following files instead? * `../tests/test_datasets/test_square.k`...

Ok perfect, you have already found the problem. I will leave this issue open. If you want/are able to propose a patch to solve this we would be more than...

Dear PR solved the problem for comma separated k files