search

microsoft / OCR-Form-Tools

A set of tools to use in Microsoft Azure Form Recognizer and OCR services.
MIT License
509 stars 171 forks source link

FOTT bug report #928

Open kirsanau opened 3 years ago

kirsanau commented 3 years ago

Time out when running layout on big files with images.

Steps to reproduce the behavior:

  1. Create project with pdf files in blob / or local file (over 40 pages with tables, and technical schemes)
  2. Click on "Run Layout on unvisited documents" when in project mode / "Run Layout" in manual mode
  3. See error

Expected behavior No time out till processing is finished or ability to configure timeout interval or ability to set pages for layout run in project mode (available in manual mode - Layout Analyze)

Screenshots file contains many pages with special charts image time out error in project mode image timeout error in manual mode (Layout analyze) image

Desktop (please complete the following information):

cannot share the file where it fails. if in manual mode i restrict number of pages for this file - it works fine.

josecu08 commented 3 years ago

Same issue here, in my case while training a custom model. It times out, however model gets correctly created after a while (checked with the API). Main problem is that when this happens the trained model doesn't appear in the application. The only way to make it work is to manually edit both the label files, so that in the manual labeling page appears as trained. Also, I have to edit the .fott file in order to be able to use the trained model within the app.