Which labelling or annoation tool you are using?

[iNeuron-ai]

May i know which is the tool being used for annotating the data?

[vsymbol]

Apply any OCR tool that help you detecting and recognizing words in the scanned document image.
For example, refer to what @4kssoft has done to the document image and generated a .json file with position and text of the image.

https://github.com/4kssoft/CUTIE/blob/master/invoice_data/Faktura1.pdf_0.json

[4kssoft]

@iNeuron-ai
#7 (comment)

[juvebogdan]

@4kssoft My training is very slow. How much time do 40k steps take? It looks like it's on CPU

[keshav-qubitrics]

@iNeuron-ai you can also try doc sumo app. Link: https://docsumo.com/

It's free to use for 100 images. You can give it a try.