This is the result of our thoughts, at Zelros, to define an open standard for a transparent use of machine learning algorithms in enterprises.
You can find:
- the initial story behind this standard in this blog post
- the explanation of the latest update in this second blog post
The standard we propose takes the form of a 'check-list', that every machine learning model embedded in a product or a service should document. We believe that filling this standard document for each AI model in production contributes to their traceability, compliance and transparency. The original standard has seven sections, explained in the blog post. NB: The new versions involve changes in the original sections.
We decided to publish this standard publicly on our github, because we know that is is only an imperfect preliminary version. It has to be reviewed by external contributors to be improved. By making our standard public, we hope it will gain visibility, and attract as much feedback as possible. We would love to learn:
- how often can the standard be used?
- in which cases is the standard not adapted?
- what is missing in the standard?
In this repository, you can find:
- the standard itself: standard.md
- an example of populated standard document, for a fake machine learning project: project-example.md
- (new) a derived standard for monitoring report: monitoring-report-standard.md
- (new) an example of populated monitoring report: monitoring-report-example.md
You can contribute by:
- suggesting PRs (pull requests)
- writing at this address