-
The stemmer has a weird behavior with quite a lot of words:
joyce -> ąć
wielce -> ąć
piwko -> ąć
royce -> ąć
pip -> ąć
To reproduce:
```
In [1]: from stempel import StempelStemmer
In …
-
## Observation
I noticed if I tried to use spacy for tokenization only (disabling 'ner', 'parser' and 'tagger') that i was getting an erroneous mapping for the english word 'spun'. As long as tagger…
-
Hi, one suggestion for the search in add-on mode.
I see from you screenshots in the description that you know German. The add-on mode works magnificently in English where you don't have a lot of su…
-
Spacy 2.1.1 - 2.1.6
Python 3.6
Ubuntu 18.04.1 LTS
In Spacy since 2.1 randomly occurs memory errors from С/С++ like "free(): invalid next size (fast)" no any additional information are provided in…
-
The current lemmatizer is a major bottleneck for the pipeline speed. We are working on porting the lemmatizer to tensorflow backend which should provide much faster prediction times.
-
## How to reproduce the behaviour
```
>>> import spacy
>>> nlp = spacy.load('es')
>>> sentence = "sencillo como habrían deseado."
>>> doc = nlp(sentence)
>>> for token in doc:
... print(tok…
-
## How to reproduce the behavior
I am not sure it's a bug, but just wanted to let you know that the lemma of the following words: 'car', 'carriers', 'scar', (SIM) card,
is always car. Is it co…
oltip updated
5 years ago
-
```
$ lsb_release -a
No LSB modules are available.
Distributor ID: Debian
Description: Debian GNU/Linux 9.9 (stretch)
Release: 9.9
Codename: stretch
$ docker --version
Docker version 19.03.0…
-
## How to reproduce the behaviour
```
>>> import spacy
>>> nlp = spacy.load('es')
>>> sentence = "No puedo hacerlo."
>>> doc = nlp(sentence)
>>> for token in doc:
... print(token.text, toke…
-
Apesar do artigo apresentar que o SentiWordNet apenas possui definição para unigramas, ainda não temos nem essa etapa pronta. Precisaríamos ter isto pronto para poder evoluir usando a estratégia com b…