MACHINE LEARNING WITH DEVOPS

Entering into the world of automation, there is a need to integrate more than one technology together. This project depicts the integration of Machine Learning with DevOps (MLOps) to train CNN Model.

In this project, I have created one Docker image using Dockerfile that creates an environment to train the respective ML Model inside Docker Container. The main aim to train the model is to attain the expected accuracy by comparing the model accuracy with the threshold value. If the model does not attain the expected accuracy then Jenkins job automatically add more dense layers in the respective position and retrain the model.

BLOCK DIAGRAM

Follow the below steps to implement the same:

Create container image that has Python3 and Keras or Numpy installed using Dockerfile.

Build the respective dockerfile using the command given below:

docker build -t mlimage:v1 .

2. Create a job chain of Job1, Job2, Job3, Job4 and Job5 using build pipeline plugin in Jenkins.

3. Job1: Pull the GitHub repository automatically when developer push local repo to GitHub.

You can view the respective Machine Learning code in the GitHub repository.

https://github.com/manishaKgupta/ML_Devops

4. Job2: By looking at the code or program file, Jenkins should automatically build the respective Docker image and launch the respective machine learning software installed container to deploy code( Example: If code uses CNN, then Jenkins should start the container that has already installed all the software required for the CNN processing).

5. Job3: Train your model and predict accuracy or metrics of the respective model.

6. Job4: If metrics accuracy is less than given threshold, then modify the code. Again re-train the model and notify that the model is trained with the expected accuracy.

7. Create one Jenkins job for monitoring i:e If container where app is running fails due to any reason then this job should automatically start the container again from where the last trained model left.

The project depicts a way to integrate Machine Learning with DevOps automation tools.

If you have any doubts or feedback, feel free to share them in the comments section below. And make sure you check out my previous blog on Automation tools that takes you from how to create a testing and production environment.

Thankyou for reading