You’re offline. This is a read only version of the page.
31
I'm trying to build an invoice processing model in AI builder that can work across multiple types of invoice from multiple vendors; at first glance, it *seems* like the solution is to create multiple collections, one for each vendor/invoice layout - then tag all my training files, and train the model.
However, I'm running into 2 major issues: for some collections, no matter how many times I correct the auto-detected fields to the appropriate data within a collection, the model never seems to retain this info; for reference, I have 10 collections, ranging from 5 to 45 documents apiece, for a total of 230 total tagged documents.
When I go to train the model, the numbers of detected fields don't match the actual numbers of fields I've tagged; for example, every single tagged document contained a correctly identified "InvoiceID" field, but on the training step, it only shows 21.
Every document within a collection contains the same fields, but not every collection does. It seems to be averaging its detection scores based on the total number of documents, and not per-collection, and it seems to be averaging its internal detection rules across all of the collections, instead of per collection.
If that's the case, what's the use in even using the collections?
Unfortunately, I was not. I ended up changing the entire project over to Syntex in Sharepoint. The cost of AI Builder credits, the lack of transparency on credit usage, and the poor performance of AI Builder overall led me to abandon using PowerPlatform for my invoice processing needs. Once again Microsoft has massively over-promised and under-delivered.
Hello, were you able to resolve the issue. I'm also facing similar issue.
I used invoices - this is specifically regarding the invoice processing model in AI Builder.
For some reason, during the “train” step of the guided workflow, the configuration seems to lose multiple instances of multiple tags across the collections used for the model. For example: I have 236 tagged documents across 11 collections, and each of them has an identified, populated, and tagged invoiceID field. So I would expect when training that the model stats would reflect 236 instances of that field. The only way to restore the missing tags is to reopen each collection, and manually click through the entire item carousel for that collection - which consumes credits. It’s clearly a backend flaw, and it’s artificially inflating my credit consumption.
#1
WarrenBelz
791
Most Valuable Professional
#2
MS.Ragavendar
410
#3
mmbr1606
275
Super User 2025 Season 1
Share
Report
All responses (
Answers (
