Be more efficient to produce ML models with mlflow

Hello, In this article I am going to make an experiment on a tool called mlflow that come out last year to help data scientist to better manage their machine learning model. The idea of this article is not to build the perfect model for the use case where I am going to build a machine learning model, but more to dive on the functionalities of mlflow and see how it can be integrated in a ML pipeline to bring efficiency in the daily basis for a data scientist/ machine learning engineer. There are three pillars around mlflow (). Their documentation is really great and they have a nice tutorial to explain the component of mlflow. For this article I am going to focus my test on the Tracking and Models parts of mlflow because I will be honest with you I didn't see the point on the Project part (looks like a conda export and a config file to run python script in a specific order) but I am sure it can help some people on the reproductive aspect of an ml pipeline.