Assigned
Status Update
No update yet.
Comments
nr...@google.com
nr...@google.com
Hello,
To assist us in conducting thorough investigation, we kindly request your cooperation in providing the following information regarding the reported issue:
- Has this scenario ever worked as expected in the past?
- Do you see this issue constantly or intermittently ?
- If this issue is seen intermittently, then how often do you observe this issue ? Is there any specific scenario or time at which this issue is observed ?
- To help us understand the issue better, please provide detailed steps to reliably reproduce the problem.
- It would be greatly helpful if you could attach screenshots of the output related to this issue.
Your cooperation in providing these details will enable us to dive deeper into the matter and work towards a prompt resolution. We appreciate your assistance and look forward to resolving this issue for you.
Thank you for your understanding and cooperation.
Description
So I am looking for workarounds to achieve this.
What you would like to accomplish:
I need a better API for training the OCR processor, so that customers can use our product to influence and train the processor themselves.
How this might work:
This can be solved in different ways:
We can collect training data from our customers in json format and transfer it to Google and further train the processor.
Or you can provide a full implementation of your interface. For example:
- upload a pdf
- identify recognisable data on it (highlight with a frame, etc.)
- send the labeling documents to Google, where they will be used for training
If applicable, reasons why alternative solutions are not sufficient:
At the moment, I'm not satisfied with the fact that the training has to be done only through the Google interface and I can't give my clients any control over the OCR they pay for. That is, we have to hire a responsible person who will train the processor ourselves, namely, sit and select data on 50 documents)
I want a little more and I think this functionality will be extremely useful and great, even if it is for an additional fee.
Other information (workarounds you have tried, documentation consulted, etc):
Now, I have only one idea. What if I first send for processing the document and then download the json file from Google cloud storage. In that json file, I can probably change the values and coordinates of the fields to the ones I need and throw them through the processor training. But it looks like an extremely troublesome crutch. Besides, I haven't succeeded yet.