Text pre-processing implemented with docker container

A simplest illustration of deploying any model (or a rule based engine in this scenario) using docker container. The main focus of this repo is to leverage Docker Images & containers to run/host the solution anywhere. This repo can be thought of as a first building block of a "production grade system" with services hosted on a cloud. Here, a text pre-processing method(can be considered as an ML model) is used, to filter only alphanumeric characters within the input text, for the simplicity and deployed the docker-image with necessary system/libraries dependencies using docker-container.

Steps

• Build the docker image (specify a different tag to improve readability -t <tag_name>)

docker build -t flask-rest-api .

• You can check and verify the docker image using images params.

docker images

• Run the docker image. Worth to note the mapping of the ports from 5000-local to 5000-docker

docker run -d -p 5000:5000 flask-rest-api

• To see that the container is in fact running:

docker ps -a

• Run unit-test using docker

docker run flask-rest-api py.test

• To show all the logs for the container

docker logs <CONTAINER ID OR CONTAINER NAME>

• Stop docker container

docker stop

For more docker related commands, follow the link below

https://docs.docker.com/reference/

Requesting to the API

curl --location --request POST 'http://0.0.0.0:5000/fetch' --header 'Content-Type: application/json' --data-raw '{"text": "what7%$$"}'