Vasiliy Alekseev

icon indicating an email [email protected]

icon location Moscow icon location Student

Results 25 comments of Vasiliy Alekseev

Seems something's wrong with PostNauka dataset encoding (if one tries to use it on Windows)

Plain `pd.read_csv()` works fine ![plain_pd](https://user-images.githubusercontent.com/15067981/82034769-4afcae80-96a7-11ea-9d97-f2012968df42.png)

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

1. Yes, my scenario was just the 1 by 1 case. Intratext coherence score cooperates with Dataset, it retrieves document texts under the hood (many documents, one by one, on...

Add some kind of Contacts section in README?

Я имел в виду bigartm в ODS. Как обычные люди про него узнают вообще?) Не стоит ли про него в Ридми заметку оставить? Насчёт контаков в целом: не знаю. У...

Add some kind of Contacts section in README?

Ещё можно в Телеграме беседу создать: наверняка порог входа в чат будет в разы меньше, чем в Слаке (где надо ещё регистрацию проходить)

Add some kind of Contacts section in README?

Канал про ТопикНет — ну ок с ним. А про tool bigartm как вообще люди узнают?) Это для меня пока загадка. Этот канал нам очень близок, и его координатами тоже...

Pip install topicnet leads to error if artm v8 already installed on Windows

Oops, installation from source also failed. Needed to remove `artm` from setup.py first, then `pip install .` did the thing

Pip install topicnet leads to error if artm v8 already installed on Windows

🤔 artm v8 really seems not compatible with topicnet > AttributeError: 'ARTM' object has no attribute 'dump_artm_model'

Pip install topicnet leads to error if artm v8 already installed on Windows

In short, it must be stated clearly that v8 ARTM is not supported, only v9 or higher)

We need to provide datasets in BOW and Natural-Order forms

Related: #57

Downloading `postnauka` dataset produces error

The trees had time to become covered with foliage and bare again. The Earth went around the Sun for millions of kilometers. The datasets were unavailable, then available, then unavailable...