Skip to main content

Notifications

Community site session details

Community site session details

Session Id : JeYVNeXFxiw2D9YNhrmVc4
Power Apps - AI Builder
Unanswered

AI Builder - Continuous Improvement for custom models

Like (0) ShareShare
ReportReport
Posted on 12 Jan 2024 16:43:29 by

Hi, I am using a custom AI model to scan documents and need to improve the model from time to time. I am not sure how the feedback loop for AI builder is meant to work.

 

The documentation recommends that AI models should be deployed in a production environment with managed solutions.

 

However the custom AI model cannot be edited in the production environment after it's been imported, so I am not quite sure how the feedback loop would be useful to improve the model in production. The documentation also mentions that feedback loop data can't transit between environments. Any thoughts on how to best use feedback loop to improve a model in the production environment after import? Thanks.

Categories:
  • cvulpe Profile Picture
    5 on 12 Nov 2024 at 09:49:23
    AI Builder - Continuous Improvement for custom models
    Hi Guys, I'm also facing a similar issue.
     
    I have a project with 40 AI models created in a default environment since there, were all the credits assigned. Now we have decided to create a designated environment and exported all the models and flows as a unmanaged solution. The strange part is that even if all the flows can be updated in this environment, for my AI models the retrained / edit feature is locked. Is there a way we can export the models from the Default environment to a production one as a editable unmanaged solution ?
     
  • samiak Profile Picture
    Moderator on 16 Jan 2024 at 14:44:25
    Re: AI Builder - Continuous Improvement for custom models

    Hi @reolan,

    Would you please tell us which custom AI model are you using?

    Samia

Under review

Thank you for your reply! To ensure a great experience for everyone, your content is awaiting approval by our Community Managers. Please check back later.

Helpful resources

Quick Links

Understanding Microsoft Agents - Introductory Session

Confused about how agents work across the Microsoft ecosystem? Register today!

Warren Belz – Community Spotlight

We are honored to recognize Warren Belz as our May 2025 Community…

Congratulations to the April Top 10 Community Stars!

Thanks for all your good work in the Community!

Leaderboard > Power Apps - AI Builder

Overall leaderboard
Loading started
Loading complete