Train & Serve a Machine Learning Model as a Microservice using Docker and Track the Model-Performance using MLflow

Let's build a recommendation engine with a collaborative filtering model, track the model results with MLflow and use Flask to serve batch-predictions of related products other users may like as well

What we will do

We will create of a local machine learning workbench aka containerized docker setup for production software-like development workflows. Having that in mind, we will use the following tools to do so:

• MLFlow for experiment tracking and model management
• MinIO to mimic AWS S3 and act as an artifact and data storage system
• PostgreSQL for a SQL engine and to serve as a backend store for MLFlow

Furthermore, we will train and track a collaborative filtering model and use the following tools for training & serving predictions of similar products other users may like as well with a RESTful API:

• implicit to train our model
• Flask REST API to serve our predictions

We will end up with an easily and quickly configurable docker setup that gets all the above tools up and running.

Data

Synthetic data created from real website data consisting of a product catalog product_catalog.csv and raw user interaction data journey.csv, needs to be copied into 0_Data. It can be accessed here .

Setting things up

For local testing purposes we can schedule our train.py script in the train container with a cronjob defined in config/cron/config.ini

[job-exec "train"]
schedule = 0 05 16 * * * 
container = recommender-local-train
command = /bin/bash -c 'pipenv run python3 /1_Train_Models/train.py'

Before building the images rename .env.example to .env. To then create all the individual images run

$ docker-compose build

To then start all the services in detached mode run

$ docker-compose up -d

All the services are now be up and running and the cronjob runs the script train.py at whatever time specified. To check running containers run

$ docker container ps

MLflow and Minio

We can access the MLflow GUI on http://localhost:5000 and the Minio Console on http://localhost:9000/ .

Testing the API

The API now runs on http://localhost:5001/related_similar_items . To test it let's run the script test_api.py (make sure you have the relevant packages installed locally)

$ python3 test_api.py`

Learn More

A more detailed explanation of the individual steps and services can be found here