How to ensure the quality of pdf uploaded to document intelliegence remains high?

Question

When uploading a pdf to the custom extraction model to train a new model, the quality of the pdf is drastically reduced compared to the original. As this affects the model's ability to read text, is there any way to make sure the quality of the pdf remains high?

Answer 1

Hello Connor,

Thanks for reaching out to us, when you mentioned "the quality of the pdf is drastically reduced compared to the original", could you please provide a sample for that so that we can look into it? Also, which API version you are working on?

As a stable solution for analyze the document you provide, I would suggest you trying GA version rather then preview version, preview version may be have some fantasy feature but there is no SLA agreement for that.

Please let me know how it works. I hope this helps.

Regards,

Yutong

-Please kindly accept the answer if you feel helpful to support the community, thanks a lot.