Vasiliy Alekseev
Vasiliy Alekseev
@karelin Thanks! We will update the notebook.
@karelin Thank you for the info! Seems we really do have some problems with the availability of the datasets :sweat_smile: I'll let you know when we fix it.
@karelin Well, it took us pretty a while to get it done) but the datasets should be available now: `demo_data = load_dataset('postnauka')` should work. It would be great to hear...
off-topic (although not quite): BaseModel has TODO in `_fit`'s docstring for `dataset_trainable` 
Exactly! The owls are not what they seem. Corrected!
> First, _fit is "protected" method, meaning we do not guarantee that it should work nice and easy for the user and that everything will work Ok, but it doesn't...
I think it is not...
Let me break into the discussion and say a couple of words in defense of *main modality* :) This is not imho about equal weights or something. Topic modeling is...
Well... yees... Ok. Then it should be clearly stated in the docstring that currently Dataset provides *no way* to modify the contents of a text collection (changing document/renaming document/adding modality/...)....