How to continuously retrain Invoices ML model with Action Center Validation Station input?

[josh.neostella (Josh Kotrous)]

I have a process that is using the Document Understanding Invoices ML Package to read data from Invoices. We are getting the Invoices sent to Action Center for human validation, and I can export the results of the human input with the Export Extraction Results activity. I also see how the ML Skill Upload File activity works to upload the export data into a dataset to then train from. How do I connect the dots between uploading the dataset with the Upload File activity, then using this in a pipeline run to train the model? Does anyone have an example they can show of this is done? I need the bot to be able to take the results from the human validation in Action Center and include these in the ML model training. Any help is much appreciated!

[system]

Hello @josh.neostella!

It seems that you have trouble getting an answer to your question in the first 24 hours.
Let us give you a few hints and helpful links.

First, make sure you browsed through our Forum FAQ Beginner’s Guide. It will teach you what should be included in your topic.

You can check out some of our resources directly, see below:

1. Always search first. It is the best way to quickly find your answer. Check out the icon for that.
   Clicking the options button will let you set more specific topic search filters, i.e. only the ones with a solution.

2. Topic that contains most common solutions with example project files can be found here.

3. Read our official documentation where you can find a lot of information and instructions about each of our products:

4. Watch the videos on our official YouTube channel for more visual tutorials.

5. Meet us and our users on our Community Slack and ask your question there.

Hopefully this will let you easily find the solution/information you need. Once you have it, we would be happy if you could share your findings here and mark it as a solution. This will help other users find it in the future.

Thank you for helping us build our UiPath Community!

Cheers from your friendly
Forum_Staff

[MartianxSpace (Chitti)]

@josh.neostella
Hi,
Welcome to the community.

Here you can find documentation on how to use Data Extraction Training: https://docs.uipath.com/document-understanding/docs/data-extraction-training-overview

[T0Bi]

I bit late but maybe I can still help you.

Here’s a new feature currently in public preview which you might want to look into:

The Auto-Fine-tuning Loop (Public Preview)

Also, I highly recommend using the Document Understanding Process Template, it can already automatically upload your trained documents:

RPA Framework for Document Understanding

Hello everyone! Our new RPA Framework for Document Understanding processes is now available for preview and review. Key features: Easy to get new Document Understanding projects started; usable in all cases - from small processes to complex solutions Easy to integrate into larger automation flows Production-ready; built-in logging, exception handling and retry mechanisms Common architecture for both Attended and Unattended (+ Action Center) implementations Meant to make develo…

Document Understanding Process - New Studio Template

Hello everyone! We are very excited to announce the new studio template for Document Understanding. You can find the Document Understanding Process template on the Official template feed - make sure Include Prerelease is checked. To get started, simply create a new project in UiPath Studio and select it. [IncludePrerelease] [image] We recommend to carefully read the enclosed User Guide, even if you’re already familiar with the solution. The Guide can be…

If this has helped you, please mark this answer as the solution.

[MartianxSpace (Chitti)]

Great News!

[aqiff]

Have you found the solution to this ? I have the same problem as you

[Pradeep.Robot (Virags)]

I believe, the training part of packaging has to be done manually for now. So, whenever you get the trained validated set out of the action center, you would require to have those outputs stored in the share drives. where as, you may train the package (retrainable one) once/thrice in a month by manually uploading the dataset in the Ai Center. I dont think there is any approach for the bot to train itself out of application outputs (Atleast for now)