Jindra Helcl

If you do this refactor, your pipeline becomes: 1. Load the trained model 2. Create the desired matrices (and find out which ones) 3. Save the model again (Note this...

I noted some situations in which tf misbehaved and the loading was failing. But it's possible that it's now solved..

This is more or less done..

Do you have an example of such newcomer? I think how the plots should look like is more of a NMT-research-related problem - how is that specific to neuralmonkey? (Just...

It's definitely a nice idea... But I don't want to do it.. :-) I think its far beyond the scope of the software (at this moment). We could do something...

Pokud by šlo ten limit nastavit jen pro konstruktory, tak bych to udělal, ale nevim, jestli to jde a jak by se to dělalo. @tomasmcz ?

Dvacet už je na refaktor zralý, to jo, ale osm je prostě málo skoro na všechno.

Hi, sorry for the late response. (Almost) all the `ini` files in the `tests` directory are working properly (those listed in the `tests/tests_run.sh` script). If you are using the master...

jak to vypadá? chystáš se k tomu?

From Python's ConfigParser documentation: > As we can see above, the API is pretty straightforward. The only bit of magic involves the DEFAULT section which provides default values for all...