kwrobel

6 ms·

1. 1.▲
   by kwrobel 7y ago

   I agree. Perplexities (probability of a text) can be compared using different tokenization after normalization.
2. 2.▲
   Personal information of 1.2bn users was left unsecured online(techradar.com)

   1 points by kwrobel 7y ago|0 comments
3. 3.▲
   by kwrobel 8y ago

   In some countries it is mandatory by law to have 14 days in one block per year.
4. 4.▲
   by kwrobel 8y ago

   "As Jeremy Howard points out, even academic papers often use softmax for multi-class classification, and I too have already seen it used incorrectly in blogs and papers during my short time studying DL." AFAIK softmax should be us
5. 5.▲
   by kwrobel 8y ago

   AFAIK in Europe you need to pay taxes only when you convert (sell) crypto currency to real currency.
6. 6.▲
   The Benefits of Attention for Document Classification(sujitpal.blogspot.com)

   2 points by kwrobel 9y ago|0 comments
7. 7.▲
   by kwrobel 9y ago

   Also in NLP domain a precision reduction may be applied: https://arxiv.org/abs/1706.06363
8. 8.▲
   by kwrobel 10y ago

   Read the paper and convince yourself.
9. 9.▲
   by kwrobel 10y ago

   Is it multi label text classification or only multi class?
10. 10.▲
    by kwrobel 10y ago

    How a sequence of words is feeded to non-recurrent network?
11. 11.▲
    by kwrobel 10y ago

    How it can be compared to keras?
12. 12.▲
    by kwrobel 11y ago

    Talking about Cyc ontology, we are working on automatic Wikipedia articles classification: http://cycloped.io/