Hendrik Kleikamp

Long short-term memory neural networks

16

This PR implements **long short-term memory** (LSTM) networks for model order reduction of instationary problems. The idea of the approach is the following: Instead of treating the time as an...

Progress bar?

19

For some tasks that take a long time to complete or that consist of many steps, a progress bar might be helpful. Always using the logging mechanism for that purpose...

Avoid copying docstrings to derived classes but refer to base class

1

See @sdrave's comment [here](https://github.com/pymor/pymor/pull/1460#discussion_r750437765).

Some small improvements of the documentation

9

This PR removes the ugly looking white background of the plots in our documentation and instead makes the background color and alpha value changeable via defaults. Further, a small mistake...

Allow `to_matrices` to accept parameter

7

For models with parameters (for instance an instationary model stemming from a PDE discretization that has been converted into an LTI system), it might be useful to be able to...

Successive constraints method

8

Since @sdrave was complaining about the missing successive constraints method for a few times now, I started implementing something in that direction. However, this is far from being finished at...

Progress bar for neural network training

3

This PR adds a progress bar using rich for the neural network training.

Thanks to @TiKeil for noticing that we still link to the old pyMOR school website.