hana9090 / fastapi_ner Goto Github PK

• My goal to deploy any ML/DL model using FastAPI and prepare a Docker to containerized the API.

Named entity recognition (NER) is one of the key information extraction tasks, which is able to identify the key elements in a text, like names of people, places, brands and more. Extracting entities from text helps in detecting the import information from unstractured data as an information extraction task.

image source

spaCy is a free open-source library for Natural Language Processing in Python. It provides multiple NLP sevices such as NER, POS tagging, word vectors and more.

FastAPI is a python web framework for building APIs and it features as a fast framework.

Docker is a tool helps developers to easily deploy and run their applications using container. Container helps in packaging the application with all dependencies of libraries to get it running.

• cd inside the app directory and run the unicorn server cd app uvicorn main:app --reload

• Go to FastAPI Swagger UI http://127.0.0.1:8000/docs to access your API

• Cd into the project directory

• Build the docker file to create the image docker build -t fastapi_ner_image .

• Create and run the container docker run -dit --name fastapi_ner_container -p 80:80 fastapi_ner_image

• Now visit the new url for the API from the running container

• • You can find it by listing the running containers with their information, such as port: docker container ls -a

• In my case it is http://0.0.0.0:80/docs Go there and test the API

video