Lemmatizing words with "'s"

[jessicabowden]

Hi,

A small bug I came across in spaCy's lemmatization when using 's in sentences.

Example 1:
In [6]: [token.lemma_ for token in en_nlp("Jane's got a new car")] Out[6]: ['jane', "'", 'get', 'a', 'new', 'car']
Here I'd expect it to extract either "Jane" or "Jane has".

In [7]: [token.lemma_ for token in en_nlp("Jane's my friend")] Out[7]: ['jane', "'s", 'my', 'friend']
Here perhaps "Jane is".

In [8]: [token.lemma_ for token in en_nlp("Jane thinks that's a nice car")] Out[8]: ['jane', 'think', 'that', "'", 'a', 'nice', 'car']
And here just an example of a non-entity token.

Thanks

[lock]

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.