How to use the SyntaxNet / tagger with the SPACy API?

Question

I used the spaCy Python package to parse and tag text and use the resulting dependency tree and other attributes to get the value. Now I would like to use SyntaxNet Parsey McParseface for parsing and binding dependencies (which seems better), but I would like to use the SPACy API because it is so easy to use and does a lot of things that Parsi doesn't have. SyntaxNet displays POS tags and dependency tags / tree in CoNLL format:

• Bob _ NOUN NNP _ 2 nsubj _ _
• brought _ VERB VBD _ 0 ROOT _ _
• _ DET DT _ 4 det _ _
• pizza _ NOUN NN _ 2 dobj _ _
• to _ ADP IN _ 2 prep _ _
• Alice _ NOUN NNP _ 5 pobj _ _
• . _., _ 2 punct _ _

and spaCy seems to be able to read the CoNLL format on the right here . But I can't figure out where the CoNLL-related string is required in the SPACy API.

Answer 1

From the spaCy blog :

Obviously, we want to build a bridge between Parsey McParseface and spaCy so you can use a more accurate model with the sweeter SPACy API.

However, it seems that a lot of work remains to be done before this is possible.

See also spaCy's answer here .

Answer 2

- SyntaxNet ? spaCy. , SyntaxNet - , .

, - CoNLL, spaCy Doc.

Answer 3

I have not tried using spaCy, but I managed to use SyntaxNet output inside Python NLTK classes / structures such as DependencyGraph and Tree.

Here is a complete example:

http://www.davidsbatista.net/blog/2017/03/25/syntaxnet/