Extend Qanary Python MT components for multiple source and target languages

qanary-component-MT-Python-HelsinkiNLP/Dockerfile

@@ -1,14 +1,21 @@
-FROM python:3.7
+FROM python:3.10
 
 COPY requirements.txt ./
 
 RUN pip install --upgrade pip
-RUN pip install -r requirements.txt; exit 0
-RUN pip install gunicorn
+RUN pip install -r requirements.txt
 
 COPY component component
+COPY utils utils
 COPY run.py boot.sh  ./
 
+# to allow preconfigured images
+ARG SOURCE_LANGUAGE
+ARG TARGET_LANGUAGE
+
+ENV SOURCE_LANGUAGE=$SOURCE_LANGUAGE
+ENV TARGET_LANGUAGE=$TARGET_LANGUAGE
+
 RUN chmod +x boot.sh
 
 ENTRYPOINT ["./boot.sh"]

qanary-component-MT-Python-HelsinkiNLP/README.md

@@ -54,8 +54,9 @@ SPRING_BOOT_ADMIN_CLIENT_INSTANCE_SERVICE-BASE-URL=http://public-component-host:
 SPRING_BOOT_ADMIN_USERNAME=admin
 SPRING_BOOT_ADMIN_PASSWORD=admin
 SERVICE_NAME_COMPONENT=MT-Helsinki-NLP
-SERVICE_DESCRIPTION_COMPONENT=Translates question to English
+SERVICE_DESCRIPTION_COMPONENT=Translates questions
 SOURCE_LANGUAGE=de
+TARGET_LANGUAGE=en
 ```
 
 The parameters description:
@@ -68,7 +69,8 @@ The parameters description:
 * `SPRING_BOOT_ADMIN_CLIENT_INSTANCE_SERVICE-BASE-URL` -- the URL of your Qanary component (has to be visible to the Qanary pipeline)
 * `SERVICE_NAME_COMPONENT` -- the name of your Qanary component (for better identification)
 * `SERVICE_DESCRIPTION_COMPONENT` -- the description of your Qanary component
-* `SOURCE_LANGUAGE` -- (optional) the source language of the text (the component will use langdetect if no source language is given)
+* `SOURCE_LANGUAGE` -- (optional) the default source language of the translation
+* `TARGET_LANGUAGE` -- (optional) the default target language of the translation 
 
 4. Build the Docker image: 
 
@@ -82,18 +84,43 @@ docker-compose build .
 docker-compose up
 ```
 
-After execution, component creates Qanary annotation in the Qanary triplestore:
+After successful execution, component creates Qanary annotation in the Qanary triplestore:
 ```
 GRAPH <uuid> {
-        ?a a qa:AnnotationOfQuestionLanguage .
-        ?a qa:translationResult "translation result" .
-        ?a qa:sourceLanguage "ISO_639-1 language code" .
-        ?a oa:annotatedBy <urn:qanary:app_name> .
-        ?a oa:annotatedAt ?time .
-    }
+  ?a a qa:AnnotationOfQuestionTranslation .
+  ?a  oa:hasTarget <urn:myQanaryQuestion> .
+  ?a  oa:hasBody "translation_result"@ISO_639-1 language code
+  ?a  oa:annotatedBy <urn:qanary:app_name> .
+  ?a  oa:annotatedAt ?time .
 }
 ```
 
+### Support for multiple Source and Target Languages
+
+This component relies on the presence of one of more existing annotations that associate a question text with a language. 
+This can be in the form of an `AnnotationOfQuestionLanguage`, as created by LD components, or an `AnnotationOfQuestionTranslation` as created by MT components.
+
+It supports multiple combinations of source and target languages. 
+You can specify a desired source and target language independently, or simply use all available language pairings. 
+
+If a `SOURCE_LANGUAGE` is set, then only texts with this specific language are considered for translation. 
+If none is set, then all configured source languages will be used to find candidates for translation. 
+
+Similarily, if a `TARGET_LANGUAGE` is set, then texts are only translated into that language. 
+If none is set, then the texts are translated into all target languages that are supported for their respective source language. 
+
+Note that while configured source languages naturally determine the possible target languages, 
+the configured target languages also determine which source languages can be supported!
+
+### Pre-configured Docker Images
+
+You may use the included file `docker-compose-pairs.yml` to build a list of images that are preconfigured for specific language pairs.
+Note that if you intend to use these containers at the same time, you need to assign different `SERVER_PORT` values for each image. 
+
+```bash
+docker-compose -f docker-compose-pairs.yml build
+```
+
 ## How To Test This Component
 
 This component uses the [pytest](https://docs.pytest.org/). 

qanary-component-MT-Python-HelsinkiNLP/boot.sh

@@ -1,16 +1,33 @@
-#!/bin/sh
+#!/bin/bash
+export $(grep -v "^#" < .env)
 
+# check required parameters
+declare -a required_vars=(
+"SPRING_BOOT_ADMIN_URL"
+"SERVER_HOST"
+"SERVER_PORT"
+"SPRING_BOOT_ADMIN_USERNAME"
+"SPRING_BOOT_ADMIN_PASSWORD"
+"SERVICE_NAME_COMPONENT"
+"SERVICE_DESCRIPTION_COMPONENT"
+)
 
-export $(grep -v '^#' .env | xargs)
+for param in ${required_vars[@]};
+do 
+    if [[ -z ${!param} ]]; then
+        echo "Required variable \"$param\" is not set!"
+        echo "The required variables are: ${required_vars[@]}"
+        exit 4
+    fi
+done
 
-echo Downloading the model
-python -c "from transformers.models.marian.modeling_marian import MarianMTModel; from transformers.models.marian.tokenization_marian import MarianTokenizer; supported_langs = ['ru', 'es', 'de', 'fr']; models = {lang: MarianMTModel.from_pretrained('Helsinki-NLP/opus-mt-{lang}-en'.format(lang=lang)) for lang in supported_langs}; tokenizers = {lang: MarianTokenizer.from_pretrained('Helsinki-NLP/opus-mt-{lang}-en'.format(lang=lang)) for lang in supported_langs}"
-echo Downloading the model finished
+echo Downloading the models
+
+python -c "from utils.model_utils import load_models_and_tokenizers; SUPPORTED_LANGS = { 'en': ['de', 'fr', 'ru', 'es'], 'de': ['en', 'fr', 'es'], 'fr': ['en', 'de', 'ru', 'es'], 'ru': ['en', 'fr', 'es'], 'es': ['en', 'de', 'fr', 'es'], }; load_models_and_tokenizers(SUPPORTED_LANGS); "
 
+echo Downloading the model finished
 
 echo The port number is: $SERVER_PORT
+echo The host is: $SERVER_HOST
 echo The Qanary pipeline URL is: $SPRING_BOOT_ADMIN_URL
-if [ -n $SERVER_PORT ]
-then
-    exec gunicorn -b :$SERVER_PORT --access-logfile - --error-logfile - run:app # refer to the gunicorn documentation for more options
-fi
+exec uvicorn run:app --host 0.0.0.0 --port $SERVER_PORT --log-level warning

qanary-component-MT-Python-HelsinkiNLP/component/__init__.py

@@ -1,27 +1,33 @@
-from component.mt_helsinki_nlp import mt_helsinki_nlp_bp
-from flask import Flask
+from component import mt_helsinki_nlp
+from fastapi import FastAPI
+from fastapi.responses import RedirectResponse, Response
 
-version = "0.1.2"
+version = "0.2.0"
 
 # default config file (use -c parameter on command line specify a custom config file)
 configfile = "app.conf"
 
 # endpoint for health information of the service required for Spring Boot Admin server callback
-healthendpoint = "/health"
-
-aboutendpoint = "/about"
+HEALTHENDPOINT = "/health"
+ABOUTENDPOINT = "/about"
+# TODO: add languages endpoint?
 
 # initialize Flask app and add the externalized service information
-app = Flask(__name__)
-app.register_blueprint(mt_helsinki_nlp_bp)
+app = FastAPI(docs_url="/swagger-ui.html")
+app.include_router(mt_helsinki_nlp.router)
+
+
+@app.get("/")
+async def main():
+    return RedirectResponse("/about")
 
 
-@app.route(healthendpoint, methods=['GET'])
+@app.get(HEALTHENDPOINT, description="Shows the status of the component")
 def health():
     """required health endpoint for callback of Spring Boot Admin server"""
-    return "alive"
+    return Response("alive", media_type="text/plain")
 
-@app.route(aboutendpoint, methods=['GET'])
+@app.get(ABOUTENDPOINT, description="Shows a description of the component")
 def about():
-    """required about endpoint for callback of Spring Boot Admin server"""
-    return "about"
+    """required about endpoint for callback of Srping Boot Admin server"""
+    return Response("Translates questions into English", media_type="text/plain")