Management is one of the most challenging task in the deployment. AWS EKS handles the management of worker nodes by themselves. Hence we can utilize our time in our business without the overhead of managing nodes. What is AWS EKS service ? Amazon Elastic Kubernetes Service (Amazon EKS) is a…

Overview: Launch an EC2 instance having the required programs(git, Apache webserver, php) installed. Enabling these services to be started on reboot. Creating an EBS volume and mount it to the EC2 instance. Creating a separate Security Group for this instance having the SSH and HTTP port exposed. Clone the github repo…

This is a deployment of an MLOPs environment in which any Convolution Neural Network(CNN) model can be checked for accuracy as per need, and will be retrained automatically by increasing no. of epochs and no. of layers. This consists of 5 independent jobs: JOB 1: Fetching the model from Github JOB 2: Launching the container with contains the required programs to train this model. JOB 3: Checking the accuracy of the model. If the accuracy is not achieved as required, this job then triggers job 4.

Normally traning a model takes huge chunk of resource and time. More so if the dataset is big as in VGG16. But with using transfer learning method, we can simply attach our dataset onto already trained model, which saves both resource and time. shHarsh/TransferLearningVGG Contribute to shHarsh/TransferLearningVGG development by creating an account on GitHub.github.com As you can see, there are 16 layers deep.

Aim: To make a fully automated deployment of project from cloning the github repo to deployment alongwith basic testing. For building this I’ve used Dockerfile method to create custom container fully capable host jenkins alongwith programs like wget, net-tools, git auto installed. I have used Centos 7 as base os…