What is the best way to train 1k different vendor invoices for custom model train? as there is a limit of 500 pages per model id
What is the best way to train a model having 1k vendor invoices, as there is a limit of 500 pages.
As per the limit mentioned we can max 100 vendor invoices per model id.
Can you please put some light on this from the implementation perspective ?

1 comment
-
Ashish Thorat commented
If we refer the pricing page for Form Recognizer, its mentioned that the limit of 500 pages per month is only for free tier.
https://azure.microsoft.com/en-us/pricing/details/cognitive-services/form-recognizer/
If your requirement has 1,000 different vendors, going with the next S0 pricing tier for Custom or Pre-built models will be the best option. And the charge will be per thousand pages that you scan.
Looking at the number of invoices that will be scanned, the price is comparatively less.
Edit 1-
Now that I read you question once again, I think you are talking about the training limitation
'The total size of the training data set must be 500 pages or less.'v2.1 which is in preview has a feature to merge multiple models. So what I think is we can create multiple models each trained with 500 pages and then merge / compose them together to create a new model.
This should solve the problem.
We might have to wait for the feature to be GA.