spaCy

mirror of https://github.com/explosion/spaCy.git synced 2025-12-31 14:03:26 +03:00

💫 Industrial-strength Natural Language Processing (NLP) in Python

ai artificial-intelligence cython data-science deep-learning entity-linking machine-learning named-entity-recognition natural-language-processing neural-network neural-networks nlp nlp-library python spacy starred-explosion-repo starred-repo text-classification tokenization

Go to file

Matthew Honnibal 7c37f45e9f * Fix unicode regex problem for non-English locales in gold standard		2016-04-28 14:31:14 +02:00
bin	* Make very hacky modifications to parser training script, to get Chinese up and running.	2016-04-28 14:30:24 +02:00
contributors
corpora/en
examples
include	pin numpy to >=1.7, ship headers	2016-04-19 19:50:42 +02:00
lang_data	* Add initial stuff for Chinese parsing	2016-04-24 18:44:24 +02:00
spacy	* Fix unicode regex problem for non-English locales in gold standard	2016-04-28 14:31:14 +02:00
website
.gitignore
.travis.yml
buildbot.json	pin numpy to >=1.7, ship headers	2016-04-19 19:50:42 +02:00
fabfile.py
LICENSE
MANIFEST.in
package.json
README.rst
requirements.txt	pin numpy to >=1.7, ship headers	2016-04-19 19:50:42 +02:00
setup.py	pin numpy to >=1.7, ship headers	2016-04-19 19:50:42 +02:00
wordnet_license.txt

README.rst

.. image:: https://travis-ci.org/spacy-io/spaCy.svg?branch=master
    :target: https://travis-ci.org/spacy-io/spaCy

==============================
spaCy: Industrial-strength NLP
==============================

spaCy is a library for advanced natural language processing in Python and Cython.

Documentation and details: https://spacy.io/

spaCy is built on the very latest research, but it isn't researchware.  It was
designed from day 1 to be used in real products. It's commercial open-source
software, released under the MIT license.


Features
--------

* Labelled dependency parsing (91.8% accuracy on OntoNotes 5)
* Named entity recognition (82.6% accuracy on OntoNotes 5)
* Part-of-speech tagging (97.1% accuracy on OntoNotes 5)
* Easy to use word vectors
* All strings mapped to integer IDs
* Export to numpy data arrays
* Alignment maintained to original string, ensuring easy mark up calculation
* Range of easy-to-use orthographic features.
* No pre-processing required. spaCy takes raw text as input, warts and newlines and all.

Top Peformance
--------------

* Fastest in the world: <50ms per document.  No faster system has ever been
  announced.
* Accuracy within 1% of the current state of the art on all tasks performed
  (parsing, named entity recognition, part-of-speech tagging).  The only more
  accurate systems are an order of magnitude slower or more.

Supports
--------

* CPython 2.6, 2.7, 3.3, 3.4, 3.5 (only 64 bit)
* OSX
* Linux
* Windows (Cygwin, MinGW, Visual Studio)