ML integration with DevOps - Current need of the industry!! A time consuming task while training the Machine Learning models is to continuously tweak the Hyper-Parameters to reach our desired Accuracy. It is one of the reasons why most of the ML related projects fail. This can be resolved upto an extent with MLOPs = ML+ DevOps In this blog, I'm explaining my MLOPs project which trains and tweaks a CNN model for Cat and Dog prediction from the dataset. My project uses "Jenkins" as an automation-tool and "GitHub" where the developer pushes the code. Requirements for setting up the project : 1. Git 2. Jenkins 3. Redhat 8 VM 4. Docker Project : Creating Environments : I've created 3 environments (images) in Docker using Dockerfile for running my programs - 1) env1 - This environment is for running any basic program which uses numpy and pandas. To run the container of env1 - docker run -it --name con_Basic env1 2) env2 -