The google normalization changes the date format in the Docuement AI (custom Invoice Parser)

[Aztec_muskan]

Under billing, we have a custom Invoice parser that has been trained, tested and deployed.  However, the parser producing incorrect date as it is auto formatted by google normalization on due_date nad invoice_date. The date in DD/MM/YYYY format use to be converted correctly to YYYY/MM/DD but is now changed to  YYYY/DD/MM by the normalizer. Correcting this automatic action is not possible and  re-training  with more data did not resolve this issue. Moreover, we cannot update the datatype or pre-process the date. Kindly help us resolve the issue.

We could not find any related information on correctly or removing the google normalization in any docs or articles available under the official documentation.

[marckevin]

Hi @Aztec_muskan,

Welcome to Google Cloud Community!

I understand you're having an issue on your Document AI custom Invoice Parser. Since Google Normalization is an automatic action, a possible solution or suggested workaround is to post-process the normalized dates, then convert to the correct Data format.

In addition, as quoted below, you can refer to this release note on Document AI for additional information on Date Normalization:

“Currently CDE Generative AI based processor supports date and currency normalization but it defaults to US date format and USD respectively in case the values are ambiguous. In other words, if a date can be parsed in mm/dd/yyyy and dd/mm/yyyy formats, it will use mm/dd/yyyy format for normalization.”

If the issue persists, I recommend reaching out to Google Cloud Support for further assistance, as they can provide insights into whether this behavior is specific to your project.

Was this helpful? If so, please accept this answer as “Solution”. If you need additional assistance, reply here within 2 business days and I’ll be happy to help.