spacy-nlp

SpaCy-1.7 python3

• Installs English tagger, parser and NER
• Installs English GloVe vectors

Getting started

• Build locally

docker build -t spacy-nlp:en .

• (OR) Get latest from hub.docker.com

docker pull pasupulaphani/spacy-nlp:en

Run tests

sh test.sh

Get shell

docker run -it spacy-nlp /bin/bash

---

spacy-nlp-zeromq

SpaCy with zeromq bindings

Getting started

• Build locally

docker build -f Dockerfile.zeromq -t spacy-nlp-zeromq:en .

• (OR) Get latest from hub.docker.com

docker pull pasupulaphani/spacy-nlp-zeromq:en

Start server

docker run --publish 4242:4242 -it spacy-nlp-zeromq:en

• (OR) Start manually

docker run -v ${PWD}:/usr/zeromq --publish 4242:4242 --entrypoint=/bin/bash  -it spacy-nlp-zeromq:en
python3 /usr/zeromq/zeromq/server.py

API

parse

$ zerorpc  tcp://0.0.0.0:4242 parse "hotel new york"

u'[{"tag": "NN", "text": "hotel new york"}]'

entities

$ zerorpc  tcp://0.0.0.0:4242 entities "hotels in london"

u'[{"end": 6, "start": 0, "text": "hotels", "type": ""}, {"end": 16, "start": 10, "text": "london", "type": ""}]'

nounChunks

$ zerorpc  tcp://0.0.0.0:4242 nounChunks "hotels in london"

u'[{"text": "hotels"}, {"text": "london"}]'

Troubleshoot

Check if port is open

if ! nc -z 0.0.0.0 4242 2>&1 >/dev/null; then echo "NOT AVAILABLE"; fi