Vasiliy Alekseev

Thetaless sometimes have troubles with modalities

2

 :cry: `np.ravel(ind > 0)` :smile: `np.ravel(ind != 0)` P.S. `0.000000001 != 0` -> Ok?

save_experiment parameter seems not working (at least in some cases)

2

```python import topicnet dataset = topicnet.cooking_machine.Dataset(...) model = topicnet.cooking_machine.models.TopicModel(...) cube = CubeCreator( num_iter=10, parameters={ "seed": [11221963], "num_topics": [5, 10] } ) exp = Experiment( model, experiment_id='exp1', save_path='should_not_be_created', save_experiment=False ) cube(model,...

Fix multiprocessing in tests

1

Currently multiprocessing may lead to the fact that tests fail from time to time. It means that builds may be red even there is all OK with the code. The...

Dataset's dictionary not updated if one changes the collection dynamically

2

* Create a `dataset` * Call `dataset.get_dictionary()` * Change dataset's `_data` by renaming one of modalities (eg. *lemmatized* -> *new_lemmatized*) * Try to build a topic model using the `dataset`...

`keep_in_memory=False` leads to the fact that `dataset.get_vw_document()` is almost unworkable

2

The method is too slow! Do we really need `dask.dataframe`? Maybe better to store documents on disk as single files (and not as one big .csv)? References: * How one...

Model's _fit should accept Dataset also, not just BatchVectorizer

6

 Seems more natural for a model to fit on Dataset. Maybe better to use `Union[artm.BatchVectorizer, topicnet.cooking_machine.Dataset]` instead of just `artm.BatchVectorizer` (Union — for compatibility)?

Add some kind of Contacts section in README?

5

How/To whom to ask a question? Maybe add info about the channel of the library in Slack?

* what is it for? * what parameters one can/is encouraged to vary using this cube (eg. number of topics maybe better to find using [OptimalNumberOfTopics](https://github.com/machine-intelligence-laboratory/OptimalNumberOfTopics) repository :slightly_smiling_face:)? * how...

Worth to note that relative weights are cool and library provides simple ways to use it. * More intuitive way to choose modality weights (not just some random values virtually...

Currently CubeCreator supports only absolute weights (am I right :slightly_smiling_face: ?). Seems that relative weights are more useful (plus taking into account that `init_simple_default_model` requires relative weights as input). +...