Abstract
We ran through in the first chapter the key themes for productionizing AI today. Before we proceed into an exhaustive look at data ingestion and techniques and tools for building an AI application, it's important to establish a framework for success.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
See also https://dataopsmanifesto.org/en/
- 2.
Source: Experian
- 3.
specifically Statistical Process Control (SPC)
- 4.
Of course some may be sandbox, proof of concept (PoC), prototype, or minimum viable product (MVP) with the low-level aims being to demonstrate/test an idea and/or demonstrate core features/user journey
- 5.
DLOps
- 6.
Google say they won’t charge unless you manually upgrade to a paid account but keep a track on usage, check API calls and free trial status (remaining credit) on the Google Console (link above) by going to Billing ➤ Overview. Remaining free trial credit is shown in the lower RHS of the screen.
- 7.
CI/CD will be discussed in the section below
- 8.
We will use the same boilerplate in Chapter 7 to build a full-stack deep learning app
- 9.
Git was originally developed in 2005 by the creator of the Linux operating system kernel
- 10.
Continuous Delivery and Continuous Deployment are similar but have slightly different goals – Continuous Deployment focuses on the end-result, that is, the actual (end-point) deployment while Continuous Delivery focuses on the process, that is, the release (steps) and release strategy
- 11.
That is, normally manual processes on GitHub such as updating code releases with notes and binary files, adding git tags to a workflow and compiling a project are all automated within Jenkins CI/CD
- 12.
So no integration to merge developer code on GitHub
- 13.
Public url is here: https://docs.docker.com/get-started
- 14.
Intended to minimize the number of defects in the quality assurance testing phase
- 15.
See Chapter 1, Important note on cloud resource usage and cost management
- 16.
Confirm and accept terms of license. If prompted by warning on authenticity, select ok. Restart Eclipse
- 17.
NB in step 2 if Maven Project is not shown, then go to Manage Jenkins ➤ Manage Plugin ➤ Available Tab. In the filter box enter “Maven plugin” and you will get search result as “Unleash Maven Plugin,” √ enable the check-box, click on “Download now and install after restart.” In the next screen click checkbox to restart Jenkins otherwise the Maven project won’t show up. You can also check this link if not seeing the Maven project: https://stackoverflow.com/questions/45205024/maven-project-option-is-not-showing-in-jenkins-under-new-item-section-latest
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to APress Media, LLC, part of Springer Nature
About this chapter
Cite this chapter
Walsh, B. (2023). AI Best Practice and DataOps. In: Productionizing AI. Apress, Berkeley, CA. https://doi.org/10.1007/978-1-4842-8817-7_2
Download citation
DOI: https://doi.org/10.1007/978-1-4842-8817-7_2
Published:
Publisher Name: Apress, Berkeley, CA
Print ISBN: 978-1-4842-8816-0
Online ISBN: 978-1-4842-8817-7
eBook Packages: Professional and Applied ComputingApress Access BooksProfessional and Applied Computing (R0)