This EMR bootstrap script installed spacy

emr-bootstrap-script-spacy.sh

#!/bin/bash -xe

#### WARNING #####
## After modifying this script you have to push it on s3 with :
## aws s3 cp emr-bootstrap-script-spacy.sh s3://tf-emr-bootstrap-sandbox-eu-west-1
## Works with EMR 5.32.0, spacy 2.3.5

version=1.1

printf "Script $version"

# Non-standard and non-Amazon Machine Image Python modules:
sudo /usr/bin/pip3.7 install -U \
  boto3 \
  pandas \
  langdetect \
  hdfs \
  tqdm \
  pathos \
  wikipedia \
  filechunkio \
  gensim \
  termcolor \
  awswrangler

# Install spacy dependancies
sudo /usr/bin/pip3.7 install -U \
  numpy \
  Cython \
  pip
# After upgrading pip, path change
sudo /usr/local/bin/pip3.7 install -U spacy

python3 -m spacy download en_core_web_sm