💫 Industrial-strength Natural Language Processing (NLP) in Python
Go to file
Matthew Honnibal d5255aad77 * Update freqs for missing tags in ner, for serializer 2015-07-23 01:17:11 +02:00
bin * Update get_freqs script 2015-07-22 15:43:06 +02:00
contributors Add CLA for suchow 2015-04-19 13:01:38 -07:00
docs * Add first draft of annotation standards doc 2015-07-14 12:50:13 +02:00
lang_data/en * Restore accidentally clobbered updates to specials.json 2015-07-20 12:19:46 +02:00
spacy * Update freqs for missing tags in ner, for serializer 2015-07-23 01:17:11 +02:00
tests * Add tests for short string optimization 2015-07-20 12:05:45 +02:00
.gitignore * Ignore spacy/serialize/*.cpp 2015-07-17 01:36:49 +02:00
.travis.yml * Have travis use pip again... 2015-06-08 01:27:08 +02:00
LICENSE.txt Tweak line spacing 2015-04-19 13:01:38 -07:00
MANIFEST.in * Add manifest file 2015-01-30 16:49:02 +11:00
README.md * Upd readme 2015-07-01 15:39:38 +02:00
bootstrap_python_env.sh * Add bootstrap script 2015-03-16 14:01:36 -04:00
dev_setup.py Tweak line spacing 2015-04-19 13:01:38 -07:00
fabfile.py * Add fab docs command 2015-07-08 12:34:35 +02:00
requirements.txt * Inc versions 2015-06-30 18:11:06 +02:00
setup.py * Add cfile.pyx 2015-07-23 01:10:36 +02:00
wordnet_license.txt * Add WordNet license file 2015-02-01 16:11:53 +11:00

README.md

spaCy: Industrial-strength NLP

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

Documentation and details: http://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. You can buy a commercial license, or you can use it under the AGPL.

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 Pefomance

  • 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 2.7
  • CPython 3.4
  • OSX
  • Linux
  • Cygwin

Want to support:

  • Visual Studio

Difficult to support:

  • PyPy 2.7
  • PyPy 3.4