Would it be awesome, if synOCR would also have the possibility to use machine learning, rather than just static tag definitions?
I am aware that this is a big Feature Request. However, it might not be necessary to completely start from scratch. I found an archived project, that could be used as reference: https://github.com/leupibr/undeep
From what I see, it seems to be an attempt of using machine learning for teaching a model and then use it for categorizing documents.
Would it be awesome, if synOCR would also have the possibility to use machine learning, rather than just static tag definitions?
I am aware that this is a big Feature Request. However, it might not be necessary to completely start from scratch. I found an archived project, that could be used as reference: https://github.com/leupibr/undeep
From what I see, it seems to be an attempt of using machine learning for teaching a model and then use it for categorizing documents.