From 8afdfb13b2d7d6f4cdd1bdd212bb0c086ed669b1 Mon Sep 17 00:00:00 2001
From: Patrick Jentsch
Date: Wed, 23 Sep 2020 15:46:43 +0200
Subject: [PATCH] Use smaller models
---
.gitlab-ci.yml | 14 +++++---------
Dockerfile | 16 ++++++++--------
nlp | 16 ++++++++--------
spacy-nlp | 16 ++++++++--------
4 files changed, 29 insertions(+), 33 deletions(-)
diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index 073b3c0..cac14f5 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -3,16 +3,12 @@ image: docker:stable
services:
- docker:stable-dind
-variables:
- DOCKER_DRIVER: overlay2
-
stages:
- build
- push
-.docker_setup:
- before_script:
- - docker login -u gitlab-ci-token -p $CI_JOB_TOKEN $CI_REGISTRY
+variables:
+ DOCKER_DRIVER: overlay2
.reg_setup:
before_script:
@@ -25,8 +21,8 @@ stages:
REG_VERSION: 0.16.1
build_image:
- extends: .docker_setup
script:
+ - docker login -u gitlab-ci-token -p $CI_JOB_TOKEN $CI_REGISTRY
- docker build -t $INTERMEDIATE_IMAGE_TAG .
- docker push $INTERMEDIATE_IMAGE_TAG
stage: build
@@ -37,11 +33,11 @@ build_image:
push_master:
extends:
- - .docker_setup
- .reg_setup
only:
- master
script:
+ - docker login -u gitlab-ci-token -p $CI_JOB_TOKEN $CI_REGISTRY
- docker pull $INTERMEDIATE_IMAGE_TAG
- /usr/local/bin/reg rm -d --auth-url $CI_REGISTRY -u $CI_REGISTRY_USER -p $CI_REGISTRY_PASSWORD $INTERMEDIATE_IMAGE_TAG
- docker tag $INTERMEDIATE_IMAGE_TAG $IMAGE_TAG
@@ -55,7 +51,6 @@ push_master:
push_other:
extends:
- - .docker_setup
- .reg_setup
except:
- master
@@ -63,6 +58,7 @@ push_other:
- branches
- tags
script:
+ - docker login -u gitlab-ci-token -p $CI_JOB_TOKEN $CI_REGISTRY
- docker pull $INTERMEDIATE_IMAGE_TAG
- /usr/local/bin/reg rm -d --auth-url $CI_REGISTRY -u $CI_REGISTRY_USER -p $CI_REGISTRY_PASSWORD $INTERMEDIATE_IMAGE_TAG
- docker tag $INTERMEDIATE_IMAGE_TAG $IMAGE_TAG
diff --git a/Dockerfile b/Dockerfile
index b090b6b..34a8e30 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -34,14 +34,14 @@ RUN apt-get install -y --no-install-recommends \
setuptools \
wheel \
&& pip3 install "spacy==${SPACY_VERSION}" \
- && python3 -m spacy download "de_core_news_lg-${SPACY_MODELS_VERSION}" --direct \
- && python3 -m spacy download "el_core_news_lg-${SPACY_MODELS_VERSION}" --direct \
- && python3 -m spacy download "en_core_web_lg-${SPACY_MODELS_VERSION}" --direct \
- && python3 -m spacy download "es_core_news_lg-${SPACY_MODELS_VERSION}" --direct \
- && python3 -m spacy download "fr_core_news_lg-${SPACY_MODELS_VERSION}" --direct \
- && python3 -m spacy download "it_core_news_lg-${SPACY_MODELS_VERSION}" --direct \
- && python3 -m spacy download "nl_core_news_lg-${SPACY_MODELS_VERSION}" --direct \
- && python3 -m spacy download "pt_core_news_lg-${SPACY_MODELS_VERSION}" --direct
+ && python3 -m spacy download "de_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+ && python3 -m spacy download "el_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+ && python3 -m spacy download "en_core_web_md-${SPACY_MODELS_VERSION}" --direct \
+ && python3 -m spacy download "es_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+ && python3 -m spacy download "fr_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+ && python3 -m spacy download "it_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+ && python3 -m spacy download "nl_core_news_md-${SPACY_MODELS_VERSION}" --direct \
+ && python3 -m spacy download "pt_core_news_md-${SPACY_MODELS_VERSION}" --direct
COPY nlp /usr/local/bin
COPY spacy-nlp /usr/local/bin
diff --git a/nlp b/nlp
index b91fb30..2bc58f8 100755
--- a/nlp
+++ b/nlp
@@ -16,14 +16,14 @@ import os
import sys
-SPACY_MODELS = {'de': 'de_core_news_sm',
- 'el': 'el_core_news_sm',
- 'en': 'en_core_web_sm',
- 'es': 'es_core_news_sm',
- 'fr': 'fr_core_news_sm',
- 'it': 'it_core_news_sm',
- 'nl': 'nl_core_news_sm',
- 'pt': 'pt_core_news_sm'}
+SPACY_MODELS = {'de': 'de_core_news_md',
+ 'el': 'el_core_news_md',
+ 'en': 'en_core_web_md',
+ 'es': 'es_core_news_md',
+ 'fr': 'fr_core_news_md',
+ 'it': 'it_core_news_md',
+ 'nl': 'nl_core_news_md',
+ 'pt': 'pt_core_news_md'}
def parse_args():
diff --git a/spacy-nlp b/spacy-nlp
index 1b75e32..de98e6b 100755
--- a/spacy-nlp
+++ b/spacy-nlp
@@ -9,14 +9,14 @@ import os
import spacy
import textwrap
-SPACY_MODELS = {'de': 'de_core_news_lg',
- 'el': 'el_core_news_lg',
- 'en': 'en_core_web_lg',
- 'es': 'es_core_news_lg',
- 'fr': 'fr_core_news_lg',
- 'it': 'it_core_news_lg',
- 'nl': 'nl_core_news_lg',
- 'pt': 'pt_core_news_lg'}
+SPACY_MODELS = {'de': 'de_core_news_md',
+ 'el': 'el_core_news_md',
+ 'en': 'en_core_web_md',
+ 'es': 'es_core_news_md',
+ 'fr': 'fr_core_news_md',
+ 'it': 'it_core_news_md',
+ 'nl': 'nl_core_news_md',
+ 'pt': 'pt_core_news_md'}
SPACY_MODELS_VERSION = os.environ.get('SPACY_MODELS_VERSION')
SPACY_VERSION = os.environ.get('SPACY_VERSION')