This paper is published in Volume-6, Issue-3, 2020
Area
Computer Science
Author
Ram Mohan Vadavalasa
Org/Univ
Independent Researcher, India
Pub. Date
26 June, 2020
Paper ID
V6I3-1581
Publisher
Keywords
Machine Learning

Citationsacebook

IEEE
Ram Mohan Vadavalasa. End to end CI/CD pipeline for Machine Learning, International Journal of Advance Research, Ideas and Innovations in Technology, www.IJARIIT.com.

APA
Ram Mohan Vadavalasa (2020). End to end CI/CD pipeline for Machine Learning. International Journal of Advance Research, Ideas and Innovations in Technology, 6(3) www.IJARIIT.com.

MLA
Ram Mohan Vadavalasa. "End to end CI/CD pipeline for Machine Learning." International Journal of Advance Research, Ideas and Innovations in Technology 6.3 (2020). www.IJARIIT.com.

Abstract

In every industry machine learning applications are becoming popular, however, compared to traditional software applications the process fo developing, deploying, and continuously improving for machine learning applications is more complex. In industry practice continuous integration, delivery, and deployment enable organizations to release new features in their products frequently. For engineering processes of developing and designing secure pipelines to support continuous practices, how machine learning systems should be architected to gain a deep understanding in the process, and how to capture, improve and report data into different aspects of continuous integration, delivery, and deployment. Without proper pipeline for machine learning it is hard to predict, test, explain, and improve data workflow behavior. Pipelining in machine learning bringing different principles and practices to machine learning applications to work in a proper manner. In the industrial sector consequences of an irregular pipeline can cause financial, resource, and time will get wasted and some times it can indirectly influence companies' personal reputation in the market. This paper discusses the problems experience while building a machine learning pipeline and ultimately describe the framework to implement the problems in the workflow. Methodically reviewing the state of the art of continuous execution to organize approaches and tools, recognize challenges and practices. As a result, the machine learning pipeline reduces the gaps and increases the speed of experimentation in the workflow.