In Classification we have different tuples. But, when it comes to filed Field Extraction Training Batches we have Vendor/Invoice/service providers folders for each one.
Problem is :
- Identification of Vendor/Invoice/service providers folders pertains to which tuple.
- Identification of Vendor/Invoice/service providers had classes or not.
Since, we don't have any fixed Vendors. We have been trained some of vendors/supplier and classification done.
But, when it comes to live , we will have some/lot of new vendors, which doesn't have classes. Hence, our auto reorganization fails.
Currently we don't have track/identification of Vendor/Invoice/service provider, which one has classes or not. Since, we have 1000's of supplier invoices and more in pipe line.
コメント
5件のコメント
Any updates please.....
Anyone found aswers.... pls
Hello,
Is it possible for you to use alternative layouts instead of Extraction Training?
Yes that also tried.. In our case some fixed elements will not recognize due to image quality and some other issues(we receive thousands of invoices and suppliers every day and we can't rectify for each and every time) ... And we don't have control on that .
If we go for extraction training 50 to 70% will recognize invoices.. Balance we can do manually.. If we can identify the which Vendor has classes or not.
Mean while we are doing layouts/Documentation also for most volume/common vendor/suppliers..
Any updates
サインインしてコメントを残してください。